docs: explain skills adoption measurement and collection

Merge pull request 'fix: canonical Target navigation and Roadmap second' (#21 ) from fix/strategy-navigation into main
fix: make Target canonical and place Roadmap second
2026-10-05 18:33:40 +00:00 · 2026-10-05 11:19:15 -07:00 · 2026-10-05 18:16:57 +00:00 · 2026-10-05 11:02:33 -07:00 · 2026-10-05 17:59:59 +00:00 · 2026-10-05 10:25:47 -07:00
29 changed files with 1021 additions and 796 deletions
@@ -2,7 +2,7 @@
  "schema_version": 1,
  "diagram_type": "architecture",
  "meta": {
-    "title": "TO-BE: Evidence before claims",
+    "title": "TO-BE: Measure skills adoption",
    "output": "static/diagrams/strategy-measurement-mobile/index.html",
    "quality_profile": "showcase",
    "legend": {
@@ -13,8 +13,8 @@
    {
      "id": "collect",
      "type": "backend",
-      "label": "Collect evidence",
+      "label": "Establish baseline",
-      "sublabel": "Task / CI / incident",
+      "sublabel": "Current workflow",
      "pos": [
        20,
        20
@@ -27,8 +27,8 @@
    {
      "id": "validate",
      "type": "backend",
-      "label": "Validate sources",
+      "label": "Join evidence",
-      "sublabel": "Joins + missingness",
+      "sublabel": "Tasks / versions / costs",
      "pos": [
        20,
        140
@@ -41,8 +41,8 @@
    {
      "id": "calculate",
      "type": "backend",
-      "label": "Calculate measures",
+      "label": "Measure the pilot",
-      "sublabel": "16 exact contracts",
+      "sublabel": "Outcomes / quality",
      "pos": [
        20,
        260
@@ -55,8 +55,8 @@
    {
      "id": "balance",
      "type": "backend",
-      "label": "Read balanced signals",
+      "label": "Report together",
-      "sublabel": "Quality / flow / risk",
+      "sublabel": "Cost / confidence / gaps",
      "pos": [
        20,
        380
@@ -69,8 +69,8 @@
    {
      "id": "weigh",
      "type": "backend",
-      "label": "Weigh benefit and cost",
+      "label": "Broader adoption",
-      "sublabel": "Outcomes / cost / people",
+      "sublabel": "Availability + use",
      "pos": [
        20,
        500
@@ -83,8 +83,8 @@
    {
      "id": "decide",
      "type": "backend",
-      "label": "Continue or change",
+      "label": "Ongoing monitoring",
-      "sublabel": "No causal ROI claim",
+      "sublabel": "Review + maintenance",
      "pos": [
        20,
        620
@@ -100,31 +100,31 @@
      "id": "flow-0",
      "from": "collect",
      "to": "validate",
-      "label": "Check"
+      "label": "prepare"
    },
    {
      "id": "flow-1",
      "from": "validate",
      "to": "calculate",
-      "label": "apply definitions"
+      "label": "collect"
    },
    {
      "id": "flow-2",
      "from": "calculate",
      "to": "balance",
-      "label": "compare"
+      "label": "calculate"
    },
    {
      "id": "flow-3",
      "from": "balance",
      "to": "weigh",
-      "label": "consider together"
+      "label": "extend"
    },
    {
      "id": "flow-4",
      "from": "weigh",
      "to": "decide",
-      "label": "decide"
+      "label": "repeat"
    }
  ]
 }
@@ -2,7 +2,7 @@
  "schema_version": 1,
  "diagram_type": "architecture",
  "meta": {
-    "title": "TO-BE: Evidence before claims",
+    "title": "TO-BE: Measure skills adoption",
    "output": "static/diagrams/strategy-measurement/index.html",
    "quality_profile": "showcase",
    "legend": {
@@ -13,8 +13,8 @@
    {
      "id": "collect",
      "type": "backend",
-      "label": "Collect evidence",
+      "label": "Establish baseline",
-      "sublabel": "Task / CI / incident",
+      "sublabel": "Current workflow",
      "pos": [
        20,
        20
@@ -27,8 +27,8 @@
    {
      "id": "validate",
      "type": "backend",
-      "label": "Validate sources",
+      "label": "Join evidence",
-      "sublabel": "Joins + missingness",
+      "sublabel": "Tasks / versions / costs",
      "pos": [
        260,
        20
@@ -41,8 +41,8 @@
    {
      "id": "calculate",
      "type": "backend",
-      "label": "Calculate measures",
+      "label": "Measure the pilot",
-      "sublabel": "16 exact contracts",
+      "sublabel": "Outcomes / quality",
      "pos": [
        260,
        150
@@ -55,8 +55,8 @@
    {
      "id": "balance",
      "type": "backend",
-      "label": "Read balanced signals",
+      "label": "Report together",
-      "sublabel": "Quality / flow / risk",
+      "sublabel": "Cost / confidence / gaps",
      "pos": [
        20,
        150
@@ -69,8 +69,8 @@
    {
      "id": "weigh",
      "type": "backend",
-      "label": "Weigh benefit and cost",
+      "label": "Broader adoption",
-      "sublabel": "Outcomes / cost / people",
+      "sublabel": "Availability + use",
      "pos": [
        20,
        280
@@ -83,8 +83,8 @@
    {
      "id": "decide",
      "type": "backend",
-      "label": "Continue or change",
+      "label": "Ongoing monitoring",
-      "sublabel": "No causal ROI claim",
+      "sublabel": "Review + maintenance",
      "pos": [
        260,
        280
@@ -100,31 +100,31 @@
      "id": "flow-0",
      "from": "collect",
      "to": "validate",
-      "label": "Check"
+      "label": "prepare"
    },
    {
      "id": "flow-1",
      "from": "validate",
      "to": "calculate",
-      "label": "apply definitions"
+      "label": "collect"
    },
    {
      "id": "flow-2",
      "from": "calculate",
      "to": "balance",
-      "label": "compare"
+      "label": "calculate"
    },
    {
      "id": "flow-3",
      "from": "balance",
      "to": "weigh",
-      "label": "consider together"
+      "label": "extend"
    },
    {
      "id": "flow-4",
      "from": "weigh",
      "to": "decide",
-      "label": "decide"
+      "label": "repeat"
    }
  ]
 }
@@ -1,156 +1,51 @@
 {
  "schema_version": 1,
  "diagram_type": "architecture",
-  "meta": {
+  "meta": {"title": "TO-BE: autonomous SDLC, human direction", "output": "static/diagrams/strategy-target-mobile/index.html", "quality_profile": "showcase", "legend": {"mode": "hidden"}},
    "title": "TO-BE: coordination without authority transfer",
    "output": "static/diagrams/strategy-target-mobile/index.html",
    "quality_profile": "showcase",
    "legend": {
      "mode": "hidden"
    }
  },
  "components": [
-    {
+    {"id":"agent","type":"security","label":"Autonomous agent","sublabel":"Human Controlled","pos":[140,200],"size":[360,120]},
-      "id": "people",
+    {"id":"scope","type":"backend","label":"Scope","sublabel":"01","pos":[100,440],"size":[240,100]},
-      "type": "backend",
+    {"id":"architecture","type":"backend","label":"Architecture","sublabel":"02","pos":[100,660],"size":[240,100]},
-      "label": "People authorize",
+    {"id":"ui-ux","type":"backend","label":"UI/UX","sublabel":"03","pos":[100,880],"size":[240,100]},
-      "sublabel": "Retain accountability",
+    {"id":"planning","type":"backend","label":"Planning","sublabel":"04","pos":[100,1100],"size":[240,100]},
-      "pos": [
+    {"id":"delivery","type":"backend","label":"Delivery","sublabel":"05","pos":[100,1320],"size":[240,100]},
-        40,
+    {"id":"uat","type":"backend","label":"UAT","sublabel":"06","pos":[100,1540],"size":[240,100]},
-        20
+    {"id":"security","type":"backend","label":"Security","sublabel":"07","pos":[100,1760],"size":[240,100]},
-      ],
+    {"id":"releases","type":"backend","label":"Releases","sublabel":"08","pos":[100,1980],"size":[240,100]},
-      "size": [
+    {"id":"code-scope","type":"frontend","label":"Code agent","sublabel":"Human-controlled","pos":[400,440],"size":[240,100]},
-        180,
+    {"id":"code-architecture","type":"frontend","label":"Code agent","sublabel":"Human-controlled","pos":[400,660],"size":[240,100]},
-        60
+    {"id":"code-ui-ux","type":"frontend","label":"Code agent","sublabel":"Human-controlled","pos":[400,880],"size":[240,100]},
-      ]
+    {"id":"code-planning","type":"frontend","label":"Code agent","sublabel":"Human-controlled","pos":[400,1100],"size":[240,100]},
-    },
+    {"id":"code-delivery","type":"frontend","label":"Code agent","sublabel":"Human-controlled","pos":[400,1320],"size":[240,100]},
-    {
+    {"id":"code-uat","type":"frontend","label":"Code agent","sublabel":"Human-controlled","pos":[400,1540],"size":[240,100]},
-      "id": "coordinator",
+    {"id":"code-security","type":"frontend","label":"Code agent","sublabel":"Human-controlled","pos":[400,1760],"size":[240,100]},
-      "type": "backend",
+    {"id":"code-releases","type":"frontend","label":"Code agent","sublabel":"Human-controlled","pos":[400,1980],"size":[240,100]}
      "label": "SDLC coordinator",
      "sublabel": "Routes approved work",
      "pos": [
        40,
        260
      ],
      "size": [
        180,
        60
      ]
    },
    {
      "id": "clients",
      "type": "backend",
      "label": "Local coding agents",
      "sublabel": "Propose changes",
      "pos": [
        40,
        380
      ],
      "size": [
        180,
        60
      ]
    },
    {
      "id": "gates",
      "type": "security",
      "label": "Repository + CI",
      "sublabel": "Verify exact revision",
      "pos": [
        40,
        500
      ],
      "size": [
        180,
        60
      ]
    },
    {
      "id": "registry",
      "type": "backend",
      "label": "Governed skills",
      "sublabel": "Pinned + qualified",
      "pos": [
        40,
        140
      ],
      "size": [
        180,
        60
      ]
    },
    {
      "id": "approval",
      "type": "security",
      "label": "Human approval",
      "sublabel": "Exact artifact + scope",
      "pos": [
        40,
        620
      ],
      "size": [
        180,
        60
      ]
    },
    {
      "id": "release",
      "type": "security",
      "label": "Release service",
      "sublabel": "Production credentials",
      "pos": [
        40,
        740
      ],
      "size": [
        180,
        60
      ]
    }
  ],
  "connections": [
-    {
+
-      "id": "flow-0",
+    {"id":"govern-scope","from":"agent","to":"scope","label":"","variant":"security","fromSide":"left","toSide":"left","via":[[60,260],[60,490]]},
-      "from": "people",
+    {"id":"govern-architecture","from":"agent","to":"architecture","label":"","variant":"security","fromSide":"left","toSide":"left","via":[[60,260],[60,710]]},
-      "to": "coordinator",
+    {"id":"govern-ui-ux","from":"agent","to":"ui-ux","label":"","variant":"security","fromSide":"left","toSide":"left","via":[[60,260],[60,930]]},
-      "label": "Authorize scope"
+    {"id":"govern-planning","from":"agent","to":"planning","label":"","variant":"security","fromSide":"left","toSide":"left","via":[[60,260],[60,1150]]},
-    },
+    {"id":"govern-delivery","from":"agent","to":"delivery","label":"","variant":"security","fromSide":"left","toSide":"left","via":[[60,260],[60,1370]]},
-    {
+    {"id":"govern-uat","from":"agent","to":"uat","label":"","variant":"security","fromSide":"left","toSide":"left","via":[[60,260],[60,1590]]},
-      "id": "flow-1",
+    {"id":"govern-security","from":"agent","to":"security","label":"","variant":"security","fromSide":"left","toSide":"left","via":[[60,260],[60,1810]]},
-      "from": "coordinator",
+    {"id":"govern-releases","from":"agent","to":"releases","label":"","variant":"security","fromSide":"left","toSide":"left","via":[[60,260],[60,2030]]},
-      "to": "clients",
+    {"id":"stage-1","from":"scope","to":"architecture","label":"Then","variant":"emphasis","fromSide":"bottom","toSide":"top"},
-      "label": "Route tasks"
+    {"id":"stage-2","from":"architecture","to":"ui-ux","label":"Then","variant":"emphasis","fromSide":"bottom","toSide":"top"},
-    },
+    {"id":"stage-3","from":"ui-ux","to":"planning","label":"Then","variant":"emphasis","fromSide":"bottom","toSide":"top"},
-    {
+    {"id":"stage-4","from":"planning","to":"delivery","label":"Then","variant":"emphasis","fromSide":"bottom","toSide":"top"},
-      "id": "flow-2",
+    {"id":"stage-5","from":"delivery","to":"uat","label":"Then","variant":"emphasis","fromSide":"bottom","toSide":"top"},
-      "from": "registry",
+    {"id":"stage-6","from":"uat","to":"security","label":"Then","variant":"emphasis","fromSide":"bottom","toSide":"top"},
-      "to": "coordinator",
+    {"id":"stage-7","from":"security","to":"releases","label":"Then","variant":"emphasis","fromSide":"bottom","toSide":"top"},
-      "label": "Guide"
+    {"id":"stage-8","from":"releases","to":"scope","label":"Repeat indefinitely","variant":"emphasis","fromSide":"bottom","toSide":"top","via":[[220,2140],[680,2140],[680,380],[220,380]],"labelAt":[440,2180]},
-    },
+    {"id":"contribute-scope","from":"code-scope","to":"scope","label":"Contribute","variant":"dashed","fromSide":"left","toSide":"right"},
-    {
+    {"id":"contribute-architecture","from":"code-architecture","to":"architecture","label":"Contribute","variant":"dashed","fromSide":"left","toSide":"right"},
-      "id": "flow-3",
+    {"id":"contribute-ui-ux","from":"code-ui-ux","to":"ui-ux","label":"Contribute","variant":"dashed","fromSide":"left","toSide":"right"},
-      "from": "registry",
+    {"id":"contribute-planning","from":"code-planning","to":"planning","label":"Contribute","variant":"dashed","fromSide":"left","toSide":"right"},
-      "to": "clients",
+    {"id":"contribute-delivery","from":"code-delivery","to":"delivery","label":"Contribute","variant":"dashed","fromSide":"left","toSide":"right"},
-      "label": "Guide"
+    {"id":"contribute-uat","from":"code-uat","to":"uat","label":"Contribute","variant":"dashed","fromSide":"left","toSide":"right"},
-    },
+    {"id":"contribute-security","from":"code-security","to":"security","label":"Contribute","variant":"dashed","fromSide":"left","toSide":"right"},
-    {
+    {"id":"contribute-releases","from":"code-releases","to":"releases","label":"Contribute","variant":"dashed","fromSide":"left","toSide":"right"}
      "id": "flow-4",
      "from": "clients",
      "to": "gates",
      "label": "Propose"
    },
    {
      "id": "flow-5",
      "from": "gates",
      "to": "approval",
      "label": "Evidence"
    },
    {
      "id": "flow-6",
      "from": "approval",
      "to": "release",
      "label": "Authorize"
    }
  ]
 }
@@ -1,156 +1,51 @@
 {
  "schema_version": 1,
  "diagram_type": "architecture",
-  "meta": {
+  "meta": {"title": "TO-BE: autonomous SDLC, human direction", "output": "static/diagrams/strategy-target/index.html", "quality_profile": "showcase", "legend": {"mode": "hidden"}},
    "title": "TO-BE: coordination without authority transfer",
    "output": "static/diagrams/strategy-target/index.html",
    "quality_profile": "showcase",
    "legend": {
      "mode": "hidden"
    }
  },
  "components": [
-    {
+    {"id":"agent","type":"security","label":"Autonomous agent","sublabel":"Human Controlled","pos":[465,450],"size":[270,90]},
-      "id": "people",
+    {"id":"scope","type":"backend","label":"Scope","sublabel":"01","pos":[510,187.5],"size":[180,75]},
-      "type": "backend",
+    {"id":"architecture","type":"backend","label":"Architecture","sublabel":"02","pos":[780,187.5],"size":[180,75]},
-      "label": "People authorize",
+    {"id":"ui-ux","type":"backend","label":"UI/UX","sublabel":"03","pos":[780,457.5],"size":[180,75]},
-      "sublabel": "Retain accountability",
+    {"id":"planning","type":"backend","label":"Planning","sublabel":"04","pos":[780,727.5],"size":[180,75]},
-      "pos": [
+    {"id":"delivery","type":"backend","label":"Delivery","sublabel":"05","pos":[510,727.5],"size":[180,75]},
-        20,
+    {"id":"uat","type":"backend","label":"UAT","sublabel":"06","pos":[240,727.5],"size":[180,75]},
-        20
+    {"id":"security","type":"backend","label":"Security","sublabel":"07","pos":[240,457.5],"size":[180,75]},
-      ],
+    {"id":"releases","type":"backend","label":"Releases","sublabel":"08","pos":[240,187.5],"size":[180,75]},
-      "size": [
+    {"id":"code-scope","type":"frontend","label":"Code agent","sublabel":"Human-controlled","pos":[510,37.5],"size":[180,75]},
-        180,
+    {"id":"code-architecture","type":"frontend","label":"Code agent","sublabel":"Human-controlled","pos":[780,37.5],"size":[180,75]},
-        60
+    {"id":"code-ui-ux","type":"frontend","label":"Code agent","sublabel":"Human-controlled","pos":[1005,457.5],"size":[180,75]},
-      ]
+    {"id":"code-planning","type":"frontend","label":"Code agent","sublabel":"Human-controlled","pos":[780,877.5],"size":[180,75]},
-    },
+    {"id":"code-delivery","type":"frontend","label":"Code agent","sublabel":"Human-controlled","pos":[510,877.5],"size":[180,75]},
-    {
+    {"id":"code-uat","type":"frontend","label":"Code agent","sublabel":"Human-controlled","pos":[240,877.5],"size":[180,75]},
-      "id": "coordinator",
+    {"id":"code-security","type":"frontend","label":"Code agent","sublabel":"Human-controlled","pos":[15,457.5],"size":[180,75]},
-      "type": "backend",
+    {"id":"code-releases","type":"frontend","label":"Code agent","sublabel":"Human-controlled","pos":[240,37.5],"size":[180,75]}
      "label": "SDLC coordinator",
      "sublabel": "Routes approved work",
      "pos": [
        20,
        140
      ],
      "size": [
        180,
        60
      ]
    },
    {
      "id": "clients",
      "type": "backend",
      "label": "Local coding agents",
      "sublabel": "Propose changes",
      "pos": [
        20,
        260
      ],
      "size": [
        180,
        60
      ]
    },
    {
      "id": "gates",
      "type": "security",
      "label": "Repository + CI",
      "sublabel": "Verify exact revision",
      "pos": [
        20,
        380
      ],
      "size": [
        180,
        60
      ]
    },
    {
      "id": "registry",
      "type": "backend",
      "label": "Governed skills",
      "sublabel": "Pinned + qualified",
      "pos": [
        260,
        140
      ],
      "size": [
        180,
        60
      ]
    },
    {
      "id": "approval",
      "type": "security",
      "label": "Human approval",
      "sublabel": "Exact artifact + scope",
      "pos": [
        260,
        380
      ],
      "size": [
        180,
        60
      ]
    },
    {
      "id": "release",
      "type": "security",
      "label": "Release service",
      "sublabel": "Production credentials",
      "pos": [
        260,
        500
      ],
      "size": [
        180,
        60
      ]
    }
  ],
  "connections": [
-    {
+
-      "id": "flow-0",
+    {"id":"govern-scope","from":"agent","to":"scope","label":"","variant":"security","fromSide":"right","toSide":"bottom","via":[[750,495],[750,292.5],[600,292.5]]},
-      "from": "people",
+    {"id":"govern-architecture","from":"agent","to":"architecture","label":"","variant":"security","fromSide":"right","toSide":"bottom","via":[[765,495],[765,345],[870,345]]},
-      "to": "coordinator",
+    {"id":"govern-ui-ux","from":"agent","to":"ui-ux","label":"","variant":"security","fromSide":"right","toSide":"left"},
-      "label": "Authorize scope"
+    {"id":"govern-planning","from":"agent","to":"planning","label":"","variant":"security","fromSide":"right","toSide":"top","via":[[765,495],[765,690],[870,690]]},
-    },
+    {"id":"govern-delivery","from":"agent","to":"delivery","label":"","variant":"security","fromSide":"bottom","toSide":"top"},
-    {
+    {"id":"govern-uat","from":"agent","to":"uat","label":"","variant":"security","fromSide":"left","toSide":"top","via":[[435,495],[435,690],[330,690]]},
-      "id": "flow-1",
+    {"id":"govern-security","from":"agent","to":"security","label":"","variant":"security","fromSide":"left","toSide":"right"},
-      "from": "coordinator",
+    {"id":"govern-releases","from":"agent","to":"releases","label":"","variant":"security","fromSide":"left","toSide":"bottom","via":[[435,495],[435,345],[330,345]]},
-      "to": "clients",
+    {"id":"stage-1","from":"scope","to":"architecture","label":"Then","variant":"emphasis","fromSide":"right","toSide":"left"},
-      "label": "Route tasks"
+    {"id":"stage-2","from":"architecture","to":"ui-ux","label":"Then","variant":"emphasis","fromSide":"bottom","toSide":"top"},
-    },
+    {"id":"stage-3","from":"ui-ux","to":"planning","label":"Then","variant":"emphasis","fromSide":"bottom","toSide":"top"},
-    {
+    {"id":"stage-4","from":"planning","to":"delivery","label":"Then","variant":"emphasis","fromSide":"left","toSide":"right"},
-      "id": "flow-2",
+    {"id":"stage-5","from":"delivery","to":"uat","label":"Then","variant":"emphasis","fromSide":"left","toSide":"right"},
-      "from": "registry",
+    {"id":"stage-6","from":"uat","to":"security","label":"Then","variant":"emphasis","fromSide":"top","toSide":"bottom"},
-      "to": "coordinator",
+    {"id":"stage-7","from":"security","to":"releases","label":"Then","variant":"emphasis","fromSide":"top","toSide":"bottom"},
-      "label": "Guide"
+    {"id":"stage-8","from":"releases","to":"scope","label":"Repeat indefinitely","variant":"emphasis","fromSide":"right","toSide":"left","labelAt":[465,170]},
-    },
+    {"id":"contribute-scope","from":"code-scope","to":"scope","label":"Contribute","variant":"dashed","fromSide":"bottom","toSide":"top"},
-    {
+    {"id":"contribute-architecture","from":"code-architecture","to":"architecture","label":"Contribute","variant":"dashed","fromSide":"bottom","toSide":"top"},
-      "id": "flow-3",
+    {"id":"contribute-ui-ux","from":"code-ui-ux","to":"ui-ux","label":"Contribute","variant":"dashed","fromSide":"left","toSide":"right","labelAt":[990,570]},
-      "from": "registry",
+    {"id":"contribute-planning","from":"code-planning","to":"planning","label":"Contribute","variant":"dashed","fromSide":"top","toSide":"bottom"},
-      "to": "clients",
+    {"id":"contribute-delivery","from":"code-delivery","to":"delivery","label":"Contribute","variant":"dashed","fromSide":"top","toSide":"bottom"},
-      "label": "Guide"
+    {"id":"contribute-uat","from":"code-uat","to":"uat","label":"Contribute","variant":"dashed","fromSide":"top","toSide":"bottom"},
-    },
+    {"id":"contribute-security","from":"code-security","to":"security","label":"Contribute","variant":"dashed","fromSide":"right","toSide":"left","labelAt":[220,570]},
-    {
+    {"id":"contribute-releases","from":"code-releases","to":"releases","label":"Contribute","variant":"dashed","fromSide":"bottom","toSide":"top"}
      "id": "flow-4",
      "from": "clients",
      "to": "gates",
      "label": "Propose"
    },
    {
      "id": "flow-5",
      "from": "gates",
      "to": "approval",
      "label": "Evidence"
    },
    {
      "id": "flow-6",
      "from": "approval",
      "to": "release",
      "label": "Authorize"
    }
  ]
 }
@@ -1,54 +0,0 @@
 ---
 id: index
 title: AI adoption strategy
 description: Govern skills first; outcomes before activity; explicit target status.
 slug: /
 hide_table_of_contents: true
 ---
 import ArchifyDiagram from '@site/src/components/ArchifyDiagram';
 <div className="strategy-content">
 **TO-BE · Target operating model — not installed controls or company results.**
 ## How do we reach dependable AI-assisted work?
 Start with governed skills, then widen coordination only when evidence supports it. The destination is one logical SDLC coordinator alongside people using qualified local coding agents.
 <ArchifyDiagram name="strategy-journey" title="Skills and quality evidence precede wider coordination" story />
 - **Skills first:** reusable working knowledge, not permission to act.
 - **People stay accountable:** the six departments and eleven teams keep their decisions.
 - **Outcomes before activity:** quality, risk and total cost determine whether to continue.
 <details>
 <summary>Strategy rationale and reading map</summary>
 **TO-BE · Target operating model.** Introduction depends on the [roadmap gates](./roadmap.mdx); these pages do not describe installed controls or measured company results.
 ### Govern skills first, expand authority only with evidence
 Eldorado's target is to make AI-assisted work dependable across the organization: useful guidance, clear human ownership, controlled execution and measurable outcomes. The first investment is a governed skills system, not an autonomous replacement for the organization. A skill packages reusable working knowledge, prerequisites and verification steps. It helps an agent perform a task; it does not grant permission or guarantee that the work is correct.
 The target combines one logical SDLC coordinating agent with developer choice on local laptops. Copilot, Claude Code and Cline remain client options within a qualified support matrix. They receive the same reviewed portable skill content through client-specific adapters. Repository, CI and release services remain the enforcement points, independently of what any agent says or which local tool produced a change.
 The strategy describes the intended operating model and the evidence required to introduce it. It does not establish that these controls are installed, that every client is supported, or that AI has improved Eldorado's performance. The current reference is the documented six-department organization. Rollout begins by validating actual clients, responsibilities, data boundaries and available evidence.
 Success means accepted outcomes with maintained quality, controlled risk and understood total cost. Faster activity without reliable outcomes is not success. More skill invocations, generated lines or tool subscriptions are not productivity targets. People remain accountable for product, customer, financial and personnel decisions.
 ## Navigate the strategy
 - [Target operating model](./target-operating-model.mdx) — One logical coordinator; local client choice; independent authority; all departments.
 - [Skills governance](./skills-governance.mdx) — Contract, lifecycle, decision rights, adapters, locking, exceptions.
 - [Skills quality](./skills-quality.mdx) — Risk tiers, independent evaluation, repeated trials, adversarial conformance.
 - [Skills adoption](./skills-adoption.mdx) — Workflow-led coaching, client choice, feedback, manual path.
 - [Measurement](./measurement.mdx) — Full KPI contracts, minimal events, illustrative calculations, privacy and limits.
 - [Skills maintenance](./skills-maintenance.mdx) — Review, drift, deprecation, rollback and incident handling.
 - [Roadmap](./roadmap.mdx) — Evidence-gated phases, no promised dates or budgets.
 </details>
 **Next:** [See how people, skills and delivery controls connect →](./target-operating-model.mdx)
 </div>
@@ -1,7 +1,8 @@
 ---
 id: measurement
-title: Measurement
+title: Skills Measurement
-description: Full KPI contracts, minimal events, illustrative calculations, privacy and limits.
+sidebar_label: Skills Measurement
 description: Which KPIs to collect during skills adoption, where the evidence comes from, how to calculate them and when to report.
 hide_table_of_contents: true
 ---
@@ -9,377 +10,62 @@ import ArchifyDiagram from '@site/src/components/ArchifyDiagram';
 <div className="strategy-content">
-**TO-BE · Target operating model — not installed controls or company results.**
+Measure adoption in stages: first establish today’s performance, then follow the pilot, extend coverage and keep skills current. Count use separately from outcomes—more use does not by itself mean better work.
-## Is AI-assisted work worth continuing?
+<ArchifyDiagram name="strategy-measurement" title="Skills adoption measurement: baseline, pilot, broader adoption and monitoring" story />
-Judge accepted outcomes alongside quality, delivery flow, risk, cost and human experience. Validate source coverage before calculating; missing evidence is unknown, not success.
+**Recommended collection plan · Target operating model, not installed telemetry.** Confirm the proposed roles and reporting cadence below before collecting data.
-<ArchifyDiagram name="strategy-measurement" title="Collect evidence, validate it, calculate balanced measures and decide" story />
+## 1. Establish the baseline
-**Illustrative only:** 1 accepted task ÷ 4 matured starts × 100 = **25%** confirmed success. One unresolved task stays in the denominator; known-outcome coverage is **75%**. These are synthetic data, not Eldorado results.
+Choose one workflow, such as reviewing a code change with a review skill. Agree what counts as a started task, independent acceptance and the follow-up period before counting outcomes. Collect current-practice evidence over a complete comparable workflow cycle and until the agreed sample requirement is met; use the same rules in the pilot.
- **16 contracts:** exact definitions below; 9 prototype calculations, 7 explicitly unimplemented metrics.
+Start with a controlled worksheet or export. Join task IDs to acceptance records, PR/commit IDs, CI results and later defects. Record the exact skill revision and client configuration for pilot work; keep missing versions unknown. Repository and CI evidence support verification, but a green build alone is not task acceptance.
 - **Balanced evidence:** no productivity rankings, composite score or causal ROI claim.
 - **Privacy first:** no raw prompts or code by default; protect small cohorts and agree retention before collection.
-<details>
+## 2. Measure the pilot
 <summary>Collection, baseline, event and privacy contracts</summary>
-**TO-BE · Target operating model.** Introduction depends on the [roadmap gates](./roadmap.mdx); these pages do not describe installed controls or measured company results.
+Start with these four questions. Capture inputs as work happens; do not reconstruct review effort from elapsed PR time.
-### Decide what counts before counting
+### Are tasks reaching acceptance?
-Use the metric contracts below for exact definitions. Every report states its eligible population, observation window, numerator and denominator or distribution sample count, source coverage, exclusions and calculation version. Zero denominators return “not available.” Missing telemetry is unknown, not inactivity, compliance or success. Examples are explicitly synthetic and illustrative.
+**Collect:** task starts and independent acceptance, failure or abandonment records from the work tracker. **Calculate:** accepted tasks / all eligible starts whose observation period has ended × 100. Failed, abandoned and unresolved tasks stay in the denominator; show unknown outcomes and coverage separately. This is [validated task success](./reference/measurement-contracts.mdx#validated-task-success), not a count of agent completions. **Weekly:** Quality Engineering reviews the pilot cohort.
-Begin with task acceptance, repository, CI, deployment and incident systems because they can independently establish outcomes. Add catalog/version inventory and task-level declarations. Client telemetry is supplementary and version-dependent; enabling rich tool details may collect sensitive commands and identifiers. Do not enable it merely to count skills. A normalized adapter may report approved skill identifiers without forwarding raw arguments.
+### Is quality holding up?
-Validated success uses all eligible initiated tasks whose predeclared observation window has matured. Failed and abandoned tasks remain nonsuccess. Unresolved or unobserved outcomes remain in the denominator, with known-outcome coverage shown alongside the conservative confirmed-success share. They are not relabeled as observed failure. A task cancelled before start is excluded; cancellation after start remains visible. Count a task once across retries and multiple skills. For timeliness, also show unfinished work and age; accepted-only lead-time distributions have survivorship limitations.
+**Collect:** accepted change IDs, linked confirmed defects, severity and follow-up dates from review and defect records; retain PR/CI evidence. **Calculate:** changes with at least one confirmed defect / accepted changes with completed follow-up × 100. Count each affected change once, not each defect. Missing links do not mean clean work. **Monthly:** Quality Engineering reviews [escaped quality](./reference/measurement-contracts.mdx#changes-with-escaped-quality-issues) after follow-up, alongside severe incident counts.
-Track all five service-level DORA measures: commit-to-production lead time, deployment frequency, change fail rate, failed deployment recovery time and deployment rework rate. Do not substitute PR-open-to-merge time or all-incident MTTR. Pair them with matured escaped-quality rates, policy effects, total cost per accepted outcome, support coverage, adoption, human experience and catalog maintenance. Keep customer/business outcomes linked to their departmental hypotheses; do not infer conversion, trust or revenue improvement from delivery telemetry.
+### What does an accepted outcome cost?
-### Baseline and interpretation
+**Collect:** invoices and usage costs plus sampled human review/rework minutes at approved rates, including unsuccessful attempts and allocated licenses/tooling. **Calculate:** total allocated workflow cost / independently accepted tasks in the same matured start cohort. Keep effort separate from waiting time. Missing buckets make this a partial estimate, not total cost. **Monthly:** Finance & Accounting reconciles [cost per accepted outcome](./reference/measurement-contracts.mdx#cost-per-accepted-outcome), restating late invoices.
-Before a pilot, freeze eligibility and acceptance rules and verify event joins. Collect prospective no-skill/current-practice evidence over a complete comparable workflow cycle and until the preregistered precision/sample requirement is met. Do not select a convenient end date after seeing results. Retain task class, service, skill, client/model configuration and concurrent changes. A matched or randomized comparison may be used when practical and ethical; label observational trends as descriptive and report confounding, selection and uncertainty.
+### Can people confidently verify the work?
-Elapsed time and human effort are different quantities. Parallel agents may reduce waiting while increasing review or compute. Include failed attempts and human review/rework in cost, do not multiply a multi-skill task's value across every invocation, and do not convert survey perceptions into proven time savings. Small or incomplete populations yield evidence limits, not an ROI percentage.
+**Collect:** an anonymous 1–5 survey after a pilot workflow cycle: verification confidence, manageable workload and usefulness as separate items. **Calculate:** sum of valid confidence scores / valid responses; also show the distribution and responses / invitations × 100. This guides enablement, not claims of saved time. **After each pilot cycle:** People & Administration reviews [confidence and cognitive load](./reference/measurement-contracts.mdx#verification-confidence-and-cognitive-load); protect small groups.
-### Event and privacy contract
+### What a pilot report looks like
-Use a versioned event envelope and required fields per event. Source producer plus event ID deduplicates transport retries. Conflicting payloads under the same key are quarantined. Attempts have distinct IDs under one task. Domain IDs deduplicate deployments, defects and approvals even when multiple feeds describe them. Late arrivals produce versioned aggregate corrections; never silently overwrite a published result. Preserve occurrence and receipt times, reject impossible orderings and flag clock skew. Missing terminal events remain unresolved until the work owner reconciles them.
+**Illustrative arithmetic, not company results or targets.** The bundled synthetic fixture shows four matured starts: one accepted, one failed, one abandoned and one unresolved. Confirmed success is **1 / 4 × 100 = 25%**, with **75%** known-outcome coverage. A newer start has not finished its observation period and is shown separately.
-Many-to-many task/attempt/skill/client associations preserve exact immutable versions. Unknown version stays null with a reason; never assign the latest revision retroactively. Client declarations, server facts and self-reports remain separately labeled. A skill-loaded event is evidence of loading only. The minimal prototype can operate without actor identity except where a lawful, necessary adoption denominator requires pseudonymous cohort membership.
+Recorded compute spend is **EUR 60 / 1 = EUR 60** per accepted outcome—a **partial estimate** because licenses, tooling, review and rework are missing. Five survey responses total 20 points: **20 / 5 = 4** out of 5; response rate is **5 / 10 × 100 = 50%**. The report calls for reconciling the unknown task and missing costs, not claiming a skills benefit. Reproduce it with `node scripts/ai-kpi.mjs`; [fixture and limits](./reference/measurement-contracts.mdx#run-the-synthetic-measurement-example).
-The privacy starting proposal is 30 days for raw structured analytics events and 12 months for aggregated trends, subject to the responsible owners' review. Keep the approved catalog/approval record for the supported revision's lifetime and its required audit period, which must be decided separately. Incident evidence uses a restricted, explicitly authorized hold and deletion schedule, not an indefinite analytics exception. Suppress reports below five people and prohibit drill-down or overlapping cohort slices that reconstruct identities. Five is a proposed minimum, not a guarantee against re-identification. No raw prompts, code, screens, customer cases or secrets are collected by default. Publish the purpose, lawful collection basis, access/deletion route and retention before collecting. Consent is not presumed to be an appropriate employment-law basis.
+## 3. Extend collection as adoption grows
-</details>
+Keep those outcome measures, then add reach. An adoption percentage is meaningless without a known eligible population.
-<details>
+**Are supported skills available?** Freeze the expected active installation inventory; compare observed bundle/adapter revisions with the approved support and revocation records. [Supported bundle coverage](./reference/measurement-contracts.mdx#supported-bundle-coverage) = fresh, verified supported installations / all expected active installations × 100. Offline or stale entries stay in the denominator. **Weekly and at revocation:** Platform & IT reconciles coverage.
 <summary>All 16 KPI contracts — formulas, populations, owners and limits</summary>
-## Metric contracts
+**Are approved workflows being used?** Freeze the lawfully enrolled participant cohort; collect workflow-linked use of an approved revision through a qualified adapter or a separately labeled declaration. [Approved-workflow adoption](./reference/measurement-contracts.mdx#approved-workflow-adoption) = eligible participants with at least one observed approved use / eligible invited participants × 100. Report this as an observed lower bound with source coverage: missing use is unknown. Installation and loading are not use. **Monthly:** Data & AI reviews reach without invocation quotas or employee rankings.
-Each contract is a prospective collection design. No live source or observed baseline is established. N and D refer to the numerator and denominator shown for that metric.
+## 4. Keep the collection running
-### Validated task success
+**Are skills still maintained?** From the versioned catalog, join each active approved revision to its owner, substantive review evidence and due date. [Review coverage](./reference/measurement-contracts.mdx#skills-within-substantive-review-window) = revisions with all three current / all active approved revisions × 100. **Monthly and each release:** Data & AI reports overdue/ownerless revisions and deprecated or revoked versions still in use—not just a percentage.
-Confirmed independently accepted tasks as a share of all eligible initiated tasks in a matured start cohort; a conservative confirmed-success share, not observed failure rate.
+At each report, compare like workflows with the baseline and publish the period, counts, exclusions and source coverage. Zero denominators mean “not available.” Trends are descriptive, not causal ROI. Act on critical policy effects immediately; add selection precision/recall at qualification and service-level delivery or departmental outcome measures when their source links are ready.
-| Contract | Definition |
+Agree purpose, lawful basis, access and retention first. No raw prompts, code or secrets by default; suppress people cohorts below five and prevent identifying drill-downs.
 | --- | --- |
 | Numerator | Distinct matured task_ids with valid independent task.accepted evidence by as_of |
 | Denominator | All distinct eligible task_ids started in window with started_at + preregistered task_observation_hours &lt;= as_of |
 | Expression | N / D * 100 |
 | Aggregation | percent |
 | Unit | % |
 | Eligible population | Tasks in a preregistered workflow cohort, started in window; each counted once across retries, skills and clients. |
 | Exclusions | Cancelled before start, sandbox/evaluation tasks and out-of-scope workflows; never post-start failures/abandonments. |
 | Events and source | task.started; task.accepted; task.failed; task.abandoned; task.cancelled; work-management acceptance record |
 | Owner / cadence | Quality Engineering / Weekly pilot; monthly departmental outcome review |
 | Missing evidence | Unresolved/missing terminal outcomes remain in denominator with unknown count and known-outcome coverage; do not label them failed. Invalid/missing start excluded from calculation but flagged as orphan evidence. |
 | Gaming and limits | Split easy tasks or omit starts; reconcile starts to approved backlog and audit acceptance rubrics. Descriptive signal, not causal AI productivity attribution. Source completeness, task mix and sampling limit inference; report uncertainty. Maturation horizon is workflow-specific and must be approved; no automatic abandonment on timeout. |
 [Full KPI reference →](./reference/measurement-contracts.mdx) preserves all definitions, collection safeguards and the distinction between synthetic calculations and unimplemented metrics.
-### Skill selection precision
+**Next:** [Maintain skills using the evidence →](./skills-maintenance.mdx)
 Correct activations among activations on an independently labeled evaluation corpus.
 | Contract | Definition |
 | --- | --- |
 | Numerator | True-positive activations for the requested workflow |
 | Denominator | True-positive plus false-positive activations |
 | Expression | N / D * 100 |
 | Aggregation | percent |
 | Unit | % |
 | Eligible population | All candidate/client labeled evaluation task-trials with known activation and relevance. |
 | Exclusions | Unlabeled/unknown activations excluded with coverage; no cherry-picked failed trials. |
 | Events and source | eval.selection; versioned labeled corpus and runner |
 | Owner / cadence | Data & AI / Each skill/client qualification |
 | Missing evidence | Unknown activation/relevance is unscorable; publish missing count and deny qualification if planned coverage incomplete. |
 | Gaming and limits | Choose easy labels or hide negative cases; freeze corpus and independently review labels. Descriptive signal, not causal AI productivity attribution. Source completeness, task mix and sampling limit inference; report uncertainty. Does not show execution success; precision depends on corpus prevalence. |
 ### Skill selection recall
 Relevant task-trials in which the intended skill activates.
 | Contract | Definition |
 | --- | --- |
 | Numerator | True-positive activations |
 | Denominator | True-positive plus false-negative task-trials |
 | Expression | N / D * 100 |
 | Aggregation | percent |
 | Unit | % |
 | Eligible population | Independently labeled relevant evaluation task-trials for a skill/client tuple. |
 | Exclusions | Irrelevant tasks; missing relevance or activation reported as unscorable, not negative. |
 | Events and source | eval.selection; versioned labeled corpus and runner |
 | Owner / cadence | Data & AI / Each skill/client qualification |
 | Missing evidence | Report evaluation coverage; null if no relevant labeled tasks. |
 | Gaming and limits | Overtrigger to maximize recall; always pair with precision and negative cases. Descriptive signal, not causal AI productivity attribution. Source completeness, task mix and sampling limit inference; report uncertainty. Loading the right skill is not adherence. |
 ### Supported bundle coverage
 Inventoried active installations with verified effective approved, supported and nonrevoked bundle/adapter tuple.
 | Contract | Definition |
 | --- | --- |
 | Numerator | Eligible installation_ids with valid inventory evidence within approved freshness window and matching support matrix |
 | Denominator | All expected active installation_ids in the managed pilot inventory |
 | Expression | N / D * 100 |
 | Aggregation | percent |
 | Unit | % |
 | Eligible population | Active installation inventory frozen for report, including invisible/offline entries until authoritative offboarding. |
 | Exclusions | Verified decommissioned installations and nonpilot assets; lack of telemetry alone never excludes. |
 | Events and source | inventory.observed; catalog.published; catalog.revoked; device inventory |
 | Owner / cadence | Platform & IT / Weekly and at revocation |
 | Missing evidence | Unknown/stale versions count outside numerator, inside denominator. Show inventory discovery coverage versus eligible device roster; if roster unknown metric is provisional. |
 | Gaming and limits | Remove silent devices or trust declared versions; reconcile managed inventory and separately label attestation trust. Descriptive signal, not causal AI productivity attribution. Source completeness, task mix and sampling limit inference; report uncertainty. Even verified file digests do not prove a private model followed the content. |
 ### Approved-workflow adoption
 Eligible participants with at least one approved-skill workflow use; usefulness assessed separately through accepted outcomes and feedback.
 | Contract | Definition |
 | --- | --- |
 | Numerator | Distinct eligible pseudonymous participants with workflow-linked skill.used on approved revision |
 | Denominator | Distinct eligible participants in frozen invited cohort |
 | Expression | N / D * 100 |
 | Aggregation | percent |
 | Unit | % |
 | Eligible population | Consented or otherwise lawfully enrolled pilot cohort for the period; supported workflow available. |
 | Exclusions | Outside pilot and formally recorded leave before period; nonresponders remain eligible, not excluded. |
 | Events and source | skill.used; cohort.snapshot; optional qualified client adapter or declaration |
 | Owner / cadence | Data & AI / Monthly |
 | Missing evidence | Show observed lower bound and telemetry/response coverage; missing usage unknown rather than non-use. |
 | Gaming and limits | Invocation quotas and coercion; no target usage count, split server/client/self-report origins. Descriptive signal, not causal AI productivity attribution. Source completeness, task mix and sampling limit inference; report uncertainty. Self-report is not verified use; installation/loading is not use or benefit. Suppress small groups. |
 ### Change lead time
 Arithmetic mean commit-to-first-successful-production-deployment duration; also publish median/p90 and sample count.
 | Contract | Definition |
 | --- | --- |
 | Numerator | Sum of elapsed hours from committed_at to first successful production deployment for each service+commit |
 | Denominator | Count of eligible service+commit pairs with valid mappings |
 | Expression | N / D |
 | Aggregation | mean |
 | Unit | hours/change |
 | Eligible population | Changes first successfully deployed to a service in reporting window; rollout retries collapse to same logical deployment. |
 | Exclusions | Nonproduction, duplicates, no-op deployment with no new commits; missing mappings shown as coverage loss. |
 | Events and source | change.committed; deployment.completed; VCS immutable commit timestamps and release manifest |
 | Owner / cadence | Platform & IT / Weekly per service |
 | Missing evidence | Missing commit mapping or negative/skewed duration excluded from numeric sample, reported explicitly; no zero substitution. |
 | Gaming and limits | Timestamp rewriting and cherry-picking fast work; lock change mapping and show undeployed backlog age. Descriptive signal, not causal AI productivity attribution. Source completeness, task mix and sampling limit inference; report uncertainty. Not PR cycle time, effort saved, or a complete view of unfinished changes. p90 uses nearest-rank sorted sample. |
 ### Deployment frequency
 Distinct successful logical production deployments per complete service-day of observation.
 | Contract | Definition |
 | --- | --- |
 | Numerator | Distinct service_id+deployment_id successful production deployments in window |
 | Denominator | Observed service-days in window |
 | Expression | N / D |
 | Aggregation | ratio |
 | Unit | deployments/service-day |
 | Eligible population | Declared production services and complete UTC observation days; report each service before pooled views. |
 | Exclusions | Test/staging, failed attempts without production effect, duplicate callbacks and retries of the same rollout. |
 | Events and source | deployment.completed; authoritative deployment service |
 | Owner / cadence | Platform & IT / Weekly per service |
 | Missing evidence | Missing feed periods not zero; report coverage and no unqualified overall rate when period incomplete. |
 | Gaming and limits | Split deployments without value; pair with success, rework and failure measures. Descriptive signal, not causal AI productivity attribution. Source completeness, task mix and sampling limit inference; report uncertainty. More deployments alone do not establish better performance; affected unsuccessful changes remain in change-fail population. |
 ### Change fail rate
 Production deployment effects requiring unplanned remediation, rollback or hotfix, not solely outage declarations.
 | Contract | Definition |
 | --- | --- |
 | Numerator | Distinct matured production deployment_ids linked to qualifying immediate intervention |
 | Denominator | All distinct production deployment_ids that reached production in the matured deployment cohort |
 | Expression | N / D * 100 |
 | Aggregation | percent |
 | Unit | % |
 | Eligible population | Deployments with production effect during window and approved intervention follow-up horizon elapsed. |
 | Exclusions | Failed preproduction attempts, duplicate callbacks and no-effect retries; failures after production effect stay included. |
 | Events and source | deployment.completed; deployment.failed; incident.linked; release and incident records |
 | Owner / cadence | Quality Engineering / Weekly; finalized after horizon |
 | Missing evidence | Known failures / all eligible deployments is a lower bound where incident linkage incomplete; label provisional and report unknown links. |
 | Gaming and limits | Avoid incident links or relabel intervention planned; reconcile rollbacks/hotfixes independently. Descriptive signal, not causal AI productivity attribution. Source completeness, task mix and sampling limit inference; report uncertainty. Define immediate intervention and follow-up horizon before pilot; no universal incident-to-AI attribution. |
 ### Failed deployment recovery time
 Mean elapsed time from production deployment failure onset to service restoration, with median/p90 and unresolved count/age.
 | Contract | Definition |
 | --- | --- |
 | Numerator | Sum of recovery_at minus failure_started_at in hours for recovered eligible deployment failures |
 | Denominator | Count of recovered eligible deployment failures with valid timestamps |
 | Expression | N / D |
 | Aggregation | mean |
 | Unit | hours/failure |
 | Eligible population | Production deployment failures beginning in window; latest verified restoration if a recovery is rescinded. |
 | Exclusions | Nondeployment incidents; invalid timestamp pairs reported missing, not treated as instantaneous recovery. |
 | Events and source | deployment.failed; service.restored; incident service |
 | Owner / cadence | Platform & IT / Weekly with unresolved failures |
 | Missing evidence | Unresolved failures reported separately and excluded only from recovered-duration distribution; prevent declaring improvement without their visibility. |
 | Gaming and limits | Premature recovery declarations or omit hard cases; require operational readback and carry unresolved ages. Descriptive signal, not causal AI productivity attribution. Source completeness, task mix and sampling limit inference; report uncertainty. Not all-incident MTTR; recovered-only sample has survivorship bias. p90 nearest rank. |
 ### Deployment rework rate
 Unplanned production deployment effects triggered by a production incident as a share of deployment effects.
 | Contract | Definition |
 | --- | --- |
 | Numerator | Distinct incident-linked deployment_ids classified unplanned remediation |
 | Denominator | All distinct production deployment_ids with production effect in window |
 | Expression | N / D * 100 |
 | Aggregation | percent |
 | Unit | % |
 | Eligible population | Production deployment effects in window with immutable planned/unplanned classification and incident linkage. |
 | Exclusions | Nonproduction, no-effect retries and duplicates; do not exclude unclassified changes silently. |
 | Events and source | deployment.completed; incident.linked; release intent classification |
 | Owner / cadence | Platform & IT / Weekly |
 | Missing evidence | Unknown classification stays denominator; observed numerator is lower bound until reconciliation; show classification coverage. |
 | Gaming and limits | Call hotfixes planned; freeze reason and independently reconcile incident/deployment links. Descriptive signal, not causal AI productivity attribution. Source completeness, task mix and sampling limit inference; report uncertainty. Distinct from failure rate: one failed deployment can lead to multiple separate rework deployments. |
 ### Changes with escaped quality issues
 Accepted changes with at least one confirmed post-acceptance defect during approved follow-up, split by severity.
 | Contract | Definition |
 | --- | --- |
 | Numerator | Distinct eligible change_ids with one or more linked confirmed defects in follow-up |
 | Denominator | Accepted change_ids with completed predeclared follow-up horizon |
 | Expression | N / D * 100 |
 | Aggregation | percent |
 | Unit | % |
 | Eligible population | Accepted changes in acceptance window whose follow-up is complete by as_of. |
 | Exclusions | Immature changes shown separately; duplicates/unconfirmed reports excluded but counted as pending triage. |
 | Events and source | change.accepted; defect.confirmed; defect tracking and acceptance evidence |
 | Owner / cadence | Quality Engineering / Monthly mature cohorts |
 | Missing evidence | Unknown linkage is a coverage warning; no fabricated clean outcome. Show absolute defect and severe incident counts as well. |
 | Gaming and limits | Shorten follow-up or downgrade severity; freeze horizon and rubric and independently sample defects. Descriptive signal, not causal AI productivity attribution. Source completeness, task mix and sampling limit inference; report uncertainty. Counts affected changes, not raw defects per change; no presumption defects are AI-caused. |
 ### Confirmed policy effects
 Eligible evaluated attempts with an actual confirmed prohibited effect; report attempts, blocked attempts and effects separately.
 | Contract | Definition |
 | --- | --- |
 | Numerator | Distinct attempt_ids with confirmed prohibited action effect |
 | Denominator | All evaluated eligible attempt_ids including failed and abandoned runs |
 | Expression | N / D * 100 |
 | Aggregation | percent |
 | Unit | % |
 | Eligible population | Governed skill trials or supported workflow attempts, reported separately for eval and live populations. |
 | Exclusions | Out-of-scope runs; detection gaps remain coverage issues, not exclusions from known initiated population. |
 | Events and source | attempt.started; policy.violation; gate.blocked; security audit and incident triage |
 | Owner / cadence | Architecture & Security / Immediate incident action; weekly aggregates |
 | Missing evidence | Unknown outcomes remain visible; zero detected effects never proves safety. Missing audits make rate provisional. |
 | Gaming and limits | Hide incidents or inflate harmless trials; absolute severity counts and fixed task mix accompany rate. Descriptive signal, not causal AI productivity attribution. Source completeness, task mix and sampling limit inference; report uncertainty. Blocked attempts are not actual effects; any critical effect blocks promotion regardless of aggregate rate. |
 ### Cost per accepted outcome
 Allocated total workflow cost including unsuccessful attempts, licenses, compute, tooling and human review/rework divided by accepted outcomes.
 | Contract | Definition |
 | --- | --- |
 | Numerator | Sum allocated costs in one approved reporting currency for all tasks/attempts in matured start cohort |
 | Denominator | Distinct independently accepted tasks in same cohort |
 | Expression | N / D |
 | Aggregation | ratio |
 | Unit | currency/accepted outcome |
 | Eligible population | Matured eligible started-task cohort, with all attempt costs through as_of and allocated shared period costs. |
 | Exclusions | Unrelated projects and documented taxes/transfers outside chosen accounting basis; never failed/abandoned attempts. |
 | Events and source | cost.recorded; task.accepted; finance invoices, usage ledger, sampled effort with approved rates |
 | Owner / cadence | Finance & Accounting Squad / Monthly with late invoice restatement |
 | Missing evidence | Missing cost buckets produce partial estimate plus coverage, not a total. Mixed currency requires dated approved conversion; zero accepted outcomes = null with full cost shown. |
 | Gaming and limits | Count tokens only or hide labor/failures; reconcile invoice and allocation totals without double counting. Descriptive signal, not causal AI productivity attribution. Source completeness, task mix and sampling limit inference; report uncertainty. Allocation and sampled labor carry uncertainty; no ROI without independently measured business benefit. |
 ### Verification confidence and cognitive load
 Three separate anonymous 1–5 item distributions: confidence in verifying output, manageable workload, usefulness. Primary item: verification confidence mean; report full distributions and response rate.
 | Contract | Definition |
 | --- | --- |
 | Numerator | Sum valid 1–5 verification-confidence responses, one per participant per survey wave |
 | Denominator | Count valid responses to that item |
 | Expression | N / D |
 | Aggregation | mean |
 | Unit | points (1–5) |
 | Eligible population | Invited eligible cohort at fixed survey wave; voluntary responses. |
 | Exclusions | Invalid/out-of-range answers and superseded duplicate submission; nonresponders stay in response-rate denominator. |
 | Events and source | survey.response; anonymous survey with single-use token and aggregate invitation count |
 | Owner / cadence | People & Administration Squad / After pilot workflow cycle; periodic follow-up |
 | Missing evidence | Report valid response count / invited count and item nonresponse; no imputation. Suppress people cohorts below five. |
 | Gaming and limits | Pressure to respond positively; no manager access to individual responses, avoid identifying free text. Descriptive signal, not causal AI productivity attribution. Source completeness, task mix and sampling limit inference; report uncertainty. Perception and selection bias; not measured time savings. No composite productivity index. |
 ### Skills within substantive review window
 Active approved skill revisions with an accountable owner and completed substantive review not past due.
 | Contract | Definition |
 | --- | --- |
 | Numerator | Distinct active approved skill revisions with current owner, review evidence and review_due_at &gt;= as_of |
 | Denominator | All active approved skill revisions in catalog snapshot |
 | Expression | N / D * 100 |
 | Aggregation | percent |
 | Unit | % |
 | Eligible population | Approved revisions currently supported for use at as_of. |
 | Exclusions | Experimental and retired revisions; deprecated/revoked in-use counts reported independently, not hidden. |
 | Events and source | catalog.published; catalog.reviewed; catalog.deprecated; catalog.revoked; version inventory |
 | Owner / cadence | Data & AI / Monthly and each release |
 | Missing evidence | Missing owner/date/evidence outside numerator; show ownerless, deprecated-in-use and revoked-in-use absolute counts. |
 | Gaming and limits | Rubber-stamp review dates; require linked regression/security/usefulness evidence. Descriptive signal, not causal AI productivity attribution. Source completeness, task mix and sampling limit inference; report uncertainty. A current review is not proof of effectiveness; consumer inventory may miss shadow copies. |
 ### Workflow-specific business acceptance
 Independently accepted domain outcomes against a predeclared business rubric, separate from technical completion.
 | Contract | Definition |
 | --- | --- |
 | Numerator | Eligible completed-observation task_ids meeting the department-specific outcome rubric |
 | Denominator | All eligible initiated workflow task_ids with matured business-outcome horizon |
 | Expression | N / D * 100 |
 | Aggregation | percent |
 | Unit | % |
 | Eligible population | One named departmental workflow and predeclared start cohort; never pool unlike rubrics. |
 | Exclusions | Cancelled-before-start and unrelated workflows; unsuccessful or abandoned eligible work remains. |
 | Events and source | task.started; business.accepted; departmental case/content/reconciliation/onboarding review record |
 | Owner / cadence | Product Management / Each pilot cycle with owning department |
 | Missing evidence | Unknown follow-up stays denominator with unresolved count; no claim actual business outcome changed without source evidence. |
 | Gaming and limits | Weaken rubric or count drafts as customer benefit; lock rubric and independently audit acceptance. Descriptive signal, not causal AI productivity attribution. Source completeness, task mix and sampling limit inference; report uncertainty. Domain owner signs acceptance: Operations service accuracy; Marketing claims/rights; Finance reconciliation accuracy; HR policy correctness; Product usability/value. Not revenue uplift. |
 </details>
 <details>
 <summary>Run the nine-metric synthetic prototype and inspect omissions</summary>
 ## Run the synthetic measurement example
 **Executed synthetic example — not Eldorado results.** The bundled fixture produces 1 independently accepted task from 4 matured starts: 25% confirmed success. One unresolved task remains in that denominator; known-outcome coverage is 75%. Selection precision and recall are each 1 / 2 = 50%. Allocated synthetic cost is EUR 60 / 1 accepted outcome = EUR 60 per accepted outcome. These deliberately small samples demonstrate arithmetic and missingness, not a baseline, target or ROI claim.
 From the private documentation checkout, run `node scripts/ai-kpi.mjs` and `node --test scripts/ai-kpi.test.mjs`. The fixture `scripts/fixtures/ai-kpi-synthetic.json` is labeled synthetic; every result carries the same label and calculation version. This prototype has no network collectors and does not enable employee telemetry.
 The executable covers validated success, selection precision/recall, change fail and rework rates, escaped quality, policy effects, cost and human experience. Other catalog metrics remain contracts rather than implemented collectors; output lists the omissions. Snapshot inputs are simplified, pre-reconciled records, not a production ingestion system. Conflicting duplicates reject the calculation rather than silently choosing a value.
 </details>
 **Next:** [Use findings to maintain, quarantine or replace skills →](./skills-maintenance.mdx)
 </div>
@@ -0,0 +1,363 @@
 ---
 id: measurement-contracts
 title: Skills KPI reference
 description: Full KPI contracts, minimal events, illustrative calculations, privacy and limits.
 ---
 <div className="strategy-content">
 Return to the [Skills Measurement collection plan](../measurement.mdx).
 **Target operating model — recommended contracts, not installed collection.** All owner and cadence entries below are proposals to confirm before the pilot, not assigned responsibilities or approved schedules.
 There are **16 contracts: 9 prototype calculations and 7 unimplemented metrics**. The prototype has no network collectors. None of these counts establishes live telemetry.
 The nine calculations cover task success, selection precision and recall, change fail rate, deployment rework, escaped quality, policy effects, cost and the primary human-experience item. The seven omissions are supported bundle coverage, approved-workflow adoption, change lead time, deployment frequency, failed deployment recovery time, substantive-review coverage and workflow-specific business acceptance.
 ## Collection and safeguards
 Introduction depends on the [roadmap gates](../roadmap.mdx); these pages do not describe installed controls or measured company results.
 ### Decide what counts before counting
 Use the metric contracts below for exact definitions. Every report states its eligible population, observation window, numerator and denominator or distribution sample count, source coverage, exclusions and calculation version. Zero denominators return “not available.” Missing telemetry is unknown, not inactivity, compliance or success. Examples are explicitly synthetic and illustrative.
 Begin with task acceptance, repository, CI, deployment and incident systems because they can independently establish outcomes. Add catalog/version inventory and task-level declarations. Client telemetry is supplementary and version-dependent; enabling rich tool details may collect sensitive commands and identifiers. Do not enable it merely to count skills. A normalized adapter may report approved skill identifiers without forwarding raw arguments.
 Validated success uses all eligible initiated tasks whose predeclared observation window has matured. Failed and abandoned tasks remain nonsuccess. Unresolved or unobserved outcomes remain in the denominator, with known-outcome coverage shown alongside the conservative confirmed-success share. They are not relabeled as observed failure. A task cancelled before start is excluded; cancellation after start remains visible. Count a task once across retries and multiple skills. For timeliness, also show unfinished work and age; accepted-only lead-time distributions have survivorship limitations.
 Track all five service-level DORA measures: commit-to-production lead time, deployment frequency, change fail rate, failed deployment recovery time and deployment rework rate. Do not substitute PR-open-to-merge time or all-incident MTTR. Pair them with matured escaped-quality rates, policy effects, total cost per accepted outcome, support coverage, adoption, human experience and catalog maintenance. Keep customer/business outcomes linked to their departmental hypotheses; do not infer conversion, trust or revenue improvement from delivery telemetry.
 ### Baseline and interpretation
 Before a pilot, freeze eligibility and acceptance rules and verify event joins. Collect prospective no-skill/current-practice evidence over a complete comparable workflow cycle and until the preregistered precision/sample requirement is met. Do not select a convenient end date after seeing results. Retain task class, service, skill, client/model configuration and concurrent changes. A matched or randomized comparison may be used when practical and ethical; label observational trends as descriptive and report confounding, selection and uncertainty.
 Elapsed time and human effort are different quantities. Parallel agents may reduce waiting while increasing review or compute. Include failed attempts and human review/rework in cost, do not multiply a multi-skill task's value across every invocation, and do not convert survey perceptions into proven time savings. Small or incomplete populations yield evidence limits, not an ROI percentage.
 ### Event and privacy contract
 Use a versioned event envelope and required fields per event. Source producer plus event ID deduplicates transport retries. Conflicting payloads under the same key are quarantined. Attempts have distinct IDs under one task. Domain IDs deduplicate deployments, defects and approvals even when multiple feeds describe them. Late arrivals produce versioned aggregate corrections; never silently overwrite a published result. Preserve occurrence and receipt times, reject impossible orderings and flag clock skew. Missing terminal events remain unresolved until the work owner reconciles them.
 Many-to-many task/attempt/skill/client associations preserve exact immutable versions. Unknown version stays null with a reason; never assign the latest revision retroactively. Client declarations, server facts and self-reports remain separately labeled. A skill-loaded event is evidence of loading only. The minimal prototype can operate without actor identity except where a lawful, necessary adoption denominator requires pseudonymous cohort membership.
 The privacy starting proposal is 30 days for raw structured analytics events and 12 months for aggregated trends, subject to the responsible owners' review. Keep the approved catalog/approval record for the supported revision's lifetime and its required audit period, which must be decided separately. Incident evidence uses a restricted, explicitly authorized hold and deletion schedule, not an indefinite analytics exception. Suppress reports below five people and prohibit drill-down or overlapping cohort slices that reconstruct identities. Five is a proposed minimum, not a guarantee against re-identification. No raw prompts, code, screens, customer cases or secrets are collected by default. Publish the purpose, lawful collection basis, access/deletion route and retention before collecting. Consent is not presumed to be an appropriate employment-law basis.
 ## Metric contracts
 Each contract is a prospective collection design. No live source or observed baseline is established. N and D refer to the numerator and denominator shown for that metric.
 ### Validated task success
 Confirmed independently accepted tasks as a share of all eligible initiated tasks in a matured start cohort; a conservative confirmed-success share, not observed failure rate.
 | Contract | Definition |
 | --- | --- |
 | Numerator | Distinct matured task_ids with valid independent task.accepted evidence by as_of |
 | Denominator | All distinct eligible task_ids started in window with started_at + preregistered task_observation_hours &lt;= as_of |
 | Expression | N / D * 100 |
 | Aggregation | percent |
 | Unit | % |
 | Eligible population | Tasks in a preregistered workflow cohort, started in window; each counted once across retries, skills and clients. |
 | Exclusions | Cancelled before start, sandbox/evaluation tasks and out-of-scope workflows; never post-start failures/abandonments. |
 | Events and source | task.started; task.accepted; task.failed; task.abandoned; task.cancelled; work-management acceptance record |
 | Owner / cadence | Quality Engineering / Weekly pilot; monthly departmental outcome review |
 | Missing evidence | Unresolved/missing terminal outcomes remain in denominator with unknown count and known-outcome coverage; do not label them failed. Invalid/missing start excluded from calculation but flagged as orphan evidence. |
 | Gaming and limits | Split easy tasks or omit starts; reconcile starts to approved backlog and audit acceptance rubrics. Descriptive signal, not causal AI productivity attribution. Source completeness, task mix and sampling limit inference; report uncertainty. Maturation horizon is workflow-specific and must be approved; no automatic abandonment on timeout. |
 ### Skill selection precision
 Correct activations among activations on an independently labeled evaluation corpus.
 | Contract | Definition |
 | --- | --- |
 | Numerator | True-positive activations for the requested workflow |
 | Denominator | True-positive plus false-positive activations |
 | Expression | N / D * 100 |
 | Aggregation | percent |
 | Unit | % |
 | Eligible population | All candidate/client labeled evaluation task-trials with known activation and relevance. |
 | Exclusions | Unlabeled/unknown activations excluded with coverage; no cherry-picked failed trials. |
 | Events and source | eval.selection; versioned labeled corpus and runner |
 | Owner / cadence | Data & AI / Each skill/client qualification |
 | Missing evidence | Unknown activation/relevance is unscorable; publish missing count and deny qualification if planned coverage incomplete. |
 | Gaming and limits | Choose easy labels or hide negative cases; freeze corpus and independently review labels. Descriptive signal, not causal AI productivity attribution. Source completeness, task mix and sampling limit inference; report uncertainty. Does not show execution success; precision depends on corpus prevalence. |
 ### Skill selection recall
 Relevant task-trials in which the intended skill activates.
 | Contract | Definition |
 | --- | --- |
 | Numerator | True-positive activations |
 | Denominator | True-positive plus false-negative task-trials |
 | Expression | N / D * 100 |
 | Aggregation | percent |
 | Unit | % |
 | Eligible population | Independently labeled relevant evaluation task-trials for a skill/client tuple. |
 | Exclusions | Irrelevant tasks; missing relevance or activation reported as unscorable, not negative. |
 | Events and source | eval.selection; versioned labeled corpus and runner |
 | Owner / cadence | Data & AI / Each skill/client qualification |
 | Missing evidence | Report evaluation coverage; null if no relevant labeled tasks. |
 | Gaming and limits | Overtrigger to maximize recall; always pair with precision and negative cases. Descriptive signal, not causal AI productivity attribution. Source completeness, task mix and sampling limit inference; report uncertainty. Loading the right skill is not adherence. |
 ### Supported bundle coverage
 Inventoried active installations with verified effective approved, supported and nonrevoked bundle/adapter tuple.
 | Contract | Definition |
 | --- | --- |
 | Numerator | Eligible installation_ids with valid inventory evidence within approved freshness window and matching support matrix |
 | Denominator | All expected active installation_ids in the managed pilot inventory |
 | Expression | N / D * 100 |
 | Aggregation | percent |
 | Unit | % |
 | Eligible population | Active installation inventory frozen for report, including invisible/offline entries until authoritative offboarding. |
 | Exclusions | Verified decommissioned installations and nonpilot assets; lack of telemetry alone never excludes. |
 | Events and source | inventory.observed; catalog.published; catalog.revoked; device inventory |
 | Owner / cadence | Platform & IT / Weekly and at revocation |
 | Missing evidence | Unknown/stale versions count outside numerator, inside denominator. Show inventory discovery coverage versus eligible device roster; if roster unknown metric is provisional. |
 | Gaming and limits | Remove silent devices or trust declared versions; reconcile managed inventory and separately label attestation trust. Descriptive signal, not causal AI productivity attribution. Source completeness, task mix and sampling limit inference; report uncertainty. Even verified file digests do not prove a private model followed the content. |
 ### Approved-workflow adoption
 Eligible participants with at least one approved-skill workflow use; usefulness assessed separately through accepted outcomes and feedback.
 | Contract | Definition |
 | --- | --- |
 | Numerator | Distinct eligible pseudonymous participants with workflow-linked skill.used on approved revision |
 | Denominator | Distinct eligible participants in frozen invited cohort |
 | Expression | N / D * 100 |
 | Aggregation | percent |
 | Unit | % |
 | Eligible population | Consented or otherwise lawfully enrolled pilot cohort for the period; supported workflow available. |
 | Exclusions | Outside pilot and formally recorded leave before period; nonresponders remain eligible, not excluded. |
 | Events and source | skill.used; cohort.snapshot; optional qualified client adapter or declaration |
 | Owner / cadence | Data & AI / Monthly |
 | Missing evidence | Show observed lower bound and telemetry/response coverage; missing usage unknown rather than non-use. |
 | Gaming and limits | Invocation quotas and coercion; no target usage count, split server/client/self-report origins. Descriptive signal, not causal AI productivity attribution. Source completeness, task mix and sampling limit inference; report uncertainty. Self-report is not verified use; installation/loading is not use or benefit. Suppress small groups. |
 ### Change lead time
 Arithmetic mean commit-to-first-successful-production-deployment duration; also publish median/p90 and sample count.
 | Contract | Definition |
 | --- | --- |
 | Numerator | Sum of elapsed hours from committed_at to first successful production deployment for each service+commit |
 | Denominator | Count of eligible service+commit pairs with valid mappings |
 | Expression | N / D |
 | Aggregation | mean |
 | Unit | hours/change |
 | Eligible population | Changes first successfully deployed to a service in reporting window; rollout retries collapse to same logical deployment. |
 | Exclusions | Nonproduction, duplicates, no-op deployment with no new commits; missing mappings shown as coverage loss. |
 | Events and source | change.committed; deployment.completed; VCS immutable commit timestamps and release manifest |
 | Owner / cadence | Platform & IT / Weekly per service |
 | Missing evidence | Missing commit mapping or negative/skewed duration excluded from numeric sample, reported explicitly; no zero substitution. |
 | Gaming and limits | Timestamp rewriting and cherry-picking fast work; lock change mapping and show undeployed backlog age. Descriptive signal, not causal AI productivity attribution. Source completeness, task mix and sampling limit inference; report uncertainty. Not PR cycle time, effort saved, or a complete view of unfinished changes. p90 uses nearest-rank sorted sample. |
 ### Deployment frequency
 Distinct successful logical production deployments per complete service-day of observation.
 | Contract | Definition |
 | --- | --- |
 | Numerator | Distinct service_id+deployment_id successful production deployments in window |
 | Denominator | Observed service-days in window |
 | Expression | N / D |
 | Aggregation | ratio |
 | Unit | deployments/service-day |
 | Eligible population | Declared production services and complete UTC observation days; report each service before pooled views. |
 | Exclusions | Test/staging, failed attempts without production effect, duplicate callbacks and retries of the same rollout. |
 | Events and source | deployment.completed; authoritative deployment service |
 | Owner / cadence | Platform & IT / Weekly per service |
 | Missing evidence | Missing feed periods not zero; report coverage and no unqualified overall rate when period incomplete. |
 | Gaming and limits | Split deployments without value; pair with success, rework and failure measures. Descriptive signal, not causal AI productivity attribution. Source completeness, task mix and sampling limit inference; report uncertainty. More deployments alone do not establish better performance; affected unsuccessful changes remain in change-fail population. |
 ### Change fail rate
 Production deployment effects requiring unplanned remediation, rollback or hotfix, not solely outage declarations.
 | Contract | Definition |
 | --- | --- |
 | Numerator | Distinct matured production deployment_ids linked to qualifying immediate intervention |
 | Denominator | All distinct production deployment_ids that reached production in the matured deployment cohort |
 | Expression | N / D * 100 |
 | Aggregation | percent |
 | Unit | % |
 | Eligible population | Deployments with production effect during window and approved intervention follow-up horizon elapsed. |
 | Exclusions | Failed preproduction attempts, duplicate callbacks and no-effect retries; failures after production effect stay included. |
 | Events and source | deployment.completed; deployment.failed; incident.linked; release and incident records |
 | Owner / cadence | Quality Engineering / Weekly; finalized after horizon |
 | Missing evidence | Known failures / all eligible deployments is a lower bound where incident linkage incomplete; label provisional and report unknown links. |
 | Gaming and limits | Avoid incident links or relabel intervention planned; reconcile rollbacks/hotfixes independently. Descriptive signal, not causal AI productivity attribution. Source completeness, task mix and sampling limit inference; report uncertainty. Define immediate intervention and follow-up horizon before pilot; no universal incident-to-AI attribution. |
 ### Failed deployment recovery time
 Mean elapsed time from production deployment failure onset to service restoration, with median/p90 and unresolved count/age.
 | Contract | Definition |
 | --- | --- |
 | Numerator | Sum of recovery_at minus failure_started_at in hours for recovered eligible deployment failures |
 | Denominator | Count of recovered eligible deployment failures with valid timestamps |
 | Expression | N / D |
 | Aggregation | mean |
 | Unit | hours/failure |
 | Eligible population | Production deployment failures beginning in window; latest verified restoration if a recovery is rescinded. |
 | Exclusions | Nondeployment incidents; invalid timestamp pairs reported missing, not treated as instantaneous recovery. |
 | Events and source | deployment.failed; service.restored; incident service |
 | Owner / cadence | Platform & IT / Weekly with unresolved failures |
 | Missing evidence | Unresolved failures reported separately and excluded only from recovered-duration distribution; prevent declaring improvement without their visibility. |
 | Gaming and limits | Premature recovery declarations or omit hard cases; require operational readback and carry unresolved ages. Descriptive signal, not causal AI productivity attribution. Source completeness, task mix and sampling limit inference; report uncertainty. Not all-incident MTTR; recovered-only sample has survivorship bias. p90 nearest rank. |
 ### Deployment rework rate
 Unplanned production deployment effects triggered by a production incident as a share of deployment effects.
 | Contract | Definition |
 | --- | --- |
 | Numerator | Distinct incident-linked deployment_ids classified unplanned remediation |
 | Denominator | All distinct production deployment_ids with production effect in window |
 | Expression | N / D * 100 |
 | Aggregation | percent |
 | Unit | % |
 | Eligible population | Production deployment effects in window with immutable planned/unplanned classification and incident linkage. |
 | Exclusions | Nonproduction, no-effect retries and duplicates; do not exclude unclassified changes silently. |
 | Events and source | deployment.completed; incident.linked; release intent classification |
 | Owner / cadence | Platform & IT / Weekly |
 | Missing evidence | Unknown classification stays denominator; observed numerator is lower bound until reconciliation; show classification coverage. |
 | Gaming and limits | Call hotfixes planned; freeze reason and independently reconcile incident/deployment links. Descriptive signal, not causal AI productivity attribution. Source completeness, task mix and sampling limit inference; report uncertainty. Distinct from failure rate: one failed deployment can lead to multiple separate rework deployments. |
 ### Changes with escaped quality issues
 Accepted changes with at least one confirmed post-acceptance defect during approved follow-up, split by severity.
 | Contract | Definition |
 | --- | --- |
 | Numerator | Distinct eligible change_ids with one or more linked confirmed defects in follow-up |
 | Denominator | Accepted change_ids with completed predeclared follow-up horizon |
 | Expression | N / D * 100 |
 | Aggregation | percent |
 | Unit | % |
 | Eligible population | Accepted changes in acceptance window whose follow-up is complete by as_of. |
 | Exclusions | Immature changes shown separately; duplicates/unconfirmed reports excluded but counted as pending triage. |
 | Events and source | change.accepted; defect.confirmed; defect tracking and acceptance evidence |
 | Owner / cadence | Quality Engineering / Monthly mature cohorts |
 | Missing evidence | Unknown linkage is a coverage warning; no fabricated clean outcome. Show absolute defect and severe incident counts as well. |
 | Gaming and limits | Shorten follow-up or downgrade severity; freeze horizon and rubric and independently sample defects. Descriptive signal, not causal AI productivity attribution. Source completeness, task mix and sampling limit inference; report uncertainty. Counts affected changes, not raw defects per change; no presumption defects are AI-caused. |
 ### Confirmed policy effects
 Eligible evaluated attempts with an actual confirmed prohibited effect; report attempts, blocked attempts and effects separately.
 | Contract | Definition |
 | --- | --- |
 | Numerator | Distinct attempt_ids with confirmed prohibited action effect |
 | Denominator | All evaluated eligible attempt_ids including failed and abandoned runs |
 | Expression | N / D * 100 |
 | Aggregation | percent |
 | Unit | % |
 | Eligible population | Governed skill trials or supported workflow attempts, reported separately for eval and live populations. |
 | Exclusions | Out-of-scope runs; detection gaps remain coverage issues, not exclusions from known initiated population. |
 | Events and source | attempt.started; policy.violation; gate.blocked; security audit and incident triage |
 | Owner / cadence | Architecture & Security / Immediate incident action; weekly aggregates |
 | Missing evidence | Unknown outcomes remain visible; zero detected effects never proves safety. Missing audits make rate provisional. |
 | Gaming and limits | Hide incidents or inflate harmless trials; absolute severity counts and fixed task mix accompany rate. Descriptive signal, not causal AI productivity attribution. Source completeness, task mix and sampling limit inference; report uncertainty. Blocked attempts are not actual effects; any critical effect blocks promotion regardless of aggregate rate. |
 ### Cost per accepted outcome
 Allocated total workflow cost including unsuccessful attempts, licenses, compute, tooling and human review/rework divided by accepted outcomes.
 | Contract | Definition |
 | --- | --- |
 | Numerator | Sum allocated costs in one approved reporting currency for all tasks/attempts in matured start cohort |
 | Denominator | Distinct independently accepted tasks in same cohort |
 | Expression | N / D |
 | Aggregation | ratio |
 | Unit | currency/accepted outcome |
 | Eligible population | Matured eligible started-task cohort, with all attempt costs through as_of and allocated shared period costs. |
 | Exclusions | Unrelated projects and documented taxes/transfers outside chosen accounting basis; never failed/abandoned attempts. |
 | Events and source | cost.recorded; task.accepted; finance invoices, usage ledger, sampled effort with approved rates |
 | Owner / cadence | Finance & Accounting Squad / Monthly with late invoice restatement |
 | Missing evidence | Missing cost buckets produce partial estimate plus coverage, not a total. Mixed currency requires dated approved conversion; zero accepted outcomes = null with full cost shown. |
 | Gaming and limits | Count tokens only or hide labor/failures; reconcile invoice and allocation totals without double counting. Descriptive signal, not causal AI productivity attribution. Source completeness, task mix and sampling limit inference; report uncertainty. Allocation and sampled labor carry uncertainty; no ROI without independently measured business benefit. |
 ### Verification confidence and cognitive load
 Three separate anonymous 1–5 item distributions: confidence in verifying output, manageable workload, usefulness. Primary item: verification confidence mean; report full distributions and response rate.
 | Contract | Definition |
 | --- | --- |
 | Numerator | Sum valid 1–5 verification-confidence responses, one per participant per survey wave |
 | Denominator | Count valid responses to that item |
 | Expression | N / D |
 | Aggregation | mean |
 | Unit | points (1–5) |
 | Eligible population | Invited eligible cohort at fixed survey wave; voluntary responses. |
 | Exclusions | Invalid/out-of-range answers and superseded duplicate submission; nonresponders stay in response-rate denominator. |
 | Events and source | survey.response; anonymous survey with single-use token and aggregate invitation count |
 | Owner / cadence | People & Administration Squad / After pilot workflow cycle; periodic follow-up |
 | Missing evidence | Report valid response count / invited count and item nonresponse; no imputation. Suppress people cohorts below five. |
 | Gaming and limits | Pressure to respond positively; no manager access to individual responses, avoid identifying free text. Descriptive signal, not causal AI productivity attribution. Source completeness, task mix and sampling limit inference; report uncertainty. Perception and selection bias; not measured time savings. No composite productivity index. |
 ### Skills within substantive review window
 Active approved skill revisions with an accountable owner and completed substantive review not past due.
 | Contract | Definition |
 | --- | --- |
 | Numerator | Distinct active approved skill revisions with current owner, review evidence and review_due_at &gt;= as_of |
 | Denominator | All active approved skill revisions in catalog snapshot |
 | Expression | N / D * 100 |
 | Aggregation | percent |
 | Unit | % |
 | Eligible population | Approved revisions currently supported for use at as_of. |
 | Exclusions | Experimental and retired revisions; deprecated/revoked in-use counts reported independently, not hidden. |
 | Events and source | catalog.published; catalog.reviewed; catalog.deprecated; catalog.revoked; version inventory |
 | Owner / cadence | Data & AI / Monthly and each release |
 | Missing evidence | Missing owner/date/evidence outside numerator; show ownerless, deprecated-in-use and revoked-in-use absolute counts. |
 | Gaming and limits | Rubber-stamp review dates; require linked regression/security/usefulness evidence. Descriptive signal, not causal AI productivity attribution. Source completeness, task mix and sampling limit inference; report uncertainty. A current review is not proof of effectiveness; consumer inventory may miss shadow copies. |
 ### Workflow-specific business acceptance
 Independently accepted domain outcomes against a predeclared business rubric, separate from technical completion.
 | Contract | Definition |
 | --- | --- |
 | Numerator | Eligible completed-observation task_ids meeting the department-specific outcome rubric |
 | Denominator | All eligible initiated workflow task_ids with matured business-outcome horizon |
 | Expression | N / D * 100 |
 | Aggregation | percent |
 | Unit | % |
 | Eligible population | One named departmental workflow and predeclared start cohort; never pool unlike rubrics. |
 | Exclusions | Cancelled-before-start and unrelated workflows; unsuccessful or abandoned eligible work remains. |
 | Events and source | task.started; business.accepted; departmental case/content/reconciliation/onboarding review record |
 | Owner / cadence | Product Management / Each pilot cycle with owning department |
 | Missing evidence | Unknown follow-up stays denominator with unresolved count; no claim actual business outcome changed without source evidence. |
 | Gaming and limits | Weaken rubric or count drafts as customer benefit; lock rubric and independently audit acceptance. Descriptive signal, not causal AI productivity attribution. Source completeness, task mix and sampling limit inference; report uncertainty. Domain owner signs acceptance: Operations service accuracy; Marketing claims/rights; Finance reconciliation accuracy; HR policy correctness; Product usability/value. Not revenue uplift. |
 ## Run the synthetic measurement example
 **Executed synthetic example — not Eldorado results.** The bundled fixture produces 1 independently accepted task from 4 matured starts: 25% confirmed success. One unresolved task remains in that denominator; known-outcome coverage is 75%. Selection precision and recall are each 1 / 2 = 50%. Recorded synthetic compute cost is EUR 60 / 1 accepted outcome = EUR 60 per accepted outcome, a partial estimate: licenses, tooling, review and rework are missing. These deliberately small samples demonstrate arithmetic and missingness, not a baseline, target or ROI claim.
 From the private documentation checkout, run `node scripts/ai-kpi.mjs` and `node --test scripts/ai-kpi.test.mjs`. The fixture `scripts/fixtures/ai-kpi-synthetic.json` is labeled synthetic; every result carries the same label and calculation version. This prototype has no network collectors and does not enable employee telemetry.
 The executable covers validated success, selection precision/recall, change fail and rework rates, escaped quality, policy effects, cost and human experience. Other catalog metrics remain contracts rather than implemented collectors; output lists the omissions. Snapshot inputs are simplified, pre-reconciled records, not a production ingestion system. Conflicting duplicates reject the calculation rather than silently choosing a value.
 **Next:** [Use findings to maintain, quarantine or replace skills →](../skills-maintenance.mdx)
 </div>
@@ -70,6 +70,6 @@ There are no invented launch dates, budgets, staffing commitments or guaranteed
 </details>
-**Next:** [Return to the strategy destination →](./index.mdx)
+**Next:** [Return to the strategy destination →](./target-operating-model.mdx)
 </div>
@@ -1,7 +1,8 @@
 ---
 id: target-operating-model
 title: Target operating model
-description: One logical coordinator; local client choice; independent authority; all departments.
+sidebar_label: Target
 description: An autonomous agent executes and governs the SDLC; humans direct, guide and review the repeating cycle.
 hide_table_of_contents: true
 ---
@@ -9,42 +10,44 @@ import ArchifyDiagram from '@site/src/components/ArchifyDiagram';
 <div className="strategy-content">
-**TO-BE · Target operating model — not installed controls or company results.**
+**TO-BE · Target operating model — the intended strategy, not a deployed system.**
-## Who coordinates, and who can authorize delivery?
+## Autonomous SDLC, human direction
-One logical coordinator routes work; people retain authority. Local agents propose changes, while independent CI and human approval govern what the release service can execute.
+<ArchifyDiagram name="strategy-target" title="Autonomous agent executes and governs the SDLC; humans direct, guide and review" story />
-<ArchifyDiagram name="strategy-target" title="People authorize; skills guide; independent gates protect release" story />
+An autonomous agent executes and governs every phase. Humans direct, guide and review it; a separate human-controlled code agent contributes to each phase from outside the cycle.
- **Local choice:** Copilot, Claude Code and Cline use qualified adapters and pinned skills.
+**Solid arrows:** lifecycle · **Dotted spokes:** execution and governance · **Dashed arrows:** contributions.
- **Central controls:** neither laptops nor the coordinator hold production credentials.
+
- **Existing organization:** six departments, eleven teams; Data &amp; AI owns coordination, not their decisions.
+**Scope → Architecture → UI/UX → Planning → Delivery → UAT → Security → Releases → Scope**
 Each stage follows the previous one. Releases returns to Scope; the cycle repeats indefinitely, with human direction informing the next iteration.
 <details>
-<summary>Trust boundaries, recovery and detailed delivery path</summary>
+<summary>Supporting implementation options: trust boundaries and recovery</summary>
-## Coordination without transferred authority
+## Controls that support autonomous execution
-The diagram is an overview; the equivalent delivery path below remains readable on narrow screens. Expand the diagram to inspect its trust boundaries and use its zoom controls for individual relationships.
+The lifecycle above is the strategic direction. The earlier control design below is a supporting implementation option, not an agreed assignment of approval roles or a claim of installed controls. Detailed approval policy and responsibilities remain to be defined with humans.
-1. Accountable people authorize the work scope; one logical coordinator routes tasks.
+1. Humans direct, guide and review; the autonomous agent executes and governs work across the lifecycle.
 2. The reviewed registry supplies pinned skills and qualified adapters to coordinator and local clients.
-3. Laptop contributors propose changes; independent repository and CI checks verify the exact revision.
+3. Laptop contributors can participate alongside the agent; independent repository and CI checks verify the exact revision.
-4. A human approves that artifact; only the release service holds production credentials.
+4. A separate release service and artifact-scoped approval are possible controls, subject to the policy humans define.
-5. Restricted audit and measurement evidence returns to people and informs the next authorized task.
+5. Restricted audit and measurement evidence supports human review and the next Scope iteration.
-<ArchifyDiagram name="ai-adoption-target" title="TO-BE AI adoption: human accountability, no direct production authority for laptops or coordinator" />
+<ArchifyDiagram name="ai-adoption-target" title="Supporting control option, not agreed approval policy: isolated release authority" />
 **TO-BE · Target operating model.** Introduction depends on the [roadmap gates](./roadmap.mdx); these pages do not describe installed controls or measured company results.
 ### One coordination function, distributed contributions
-Data & AI owns the coordinating function, with the existing AI Architect function connecting it to Technology leadership. The logical coordinator maintains approved work records, dependencies, evidence requests, bounded retries and escalation. It can summarize, propose work and route decisions. It cannot declare its own work approved, change protected checks, authorize production release or replace the responsible human.
+The autonomous agent is more than a task router: it executes and governs the repeating SDLC under human direction, guidance and review. A coordinating function maintains work records, dependencies, evidence requests, bounded retries and escalation. Data & AI and the existing AI Architect function are possible implementation contributors, not newly assigned approval authorities. Autonomous execution does not remove human involvement or allow the agent to redefine its own controls.
 One logical coordinator does not mean one fragile process. A durable task ledger, explicit state transitions, leases and idempotency keys allow recovery or multiple replicas without duplicate delivery. Inference outages stop new agent decisions, not the organization's ability to take over. People can resume work through the normal deterministic workflow. Evidence and authorization services fail closed for privileged actions; missing evidence never becomes permission.
-The delivery path is: an approved problem and acceptance criteria; an authorized task; local or isolated worker proposals; independently reviewed repository changes; exact-revision CI evidence; a human release decision; execution by the release service; then operational feedback. Product Management accepts the outcome hypothesis and business result. Product Design reviews experience and accessibility. Product Engineering owns the code and maintainability. Quality Engineering assesses quality independently. Architecture & Security assesses risk. Platform & IT operates the delivery boundary. No arrow from a laptop or coordinator goes directly to production authority.
+These controls support the sequence Scope → Architecture → UI/UX → Planning → Delivery → UAT → Security → Releases → Scope; they do not replace it with a separate lifecycle. Product, engineering, design, quality, security and platform expertise inform human direction and review. Specific approval assignments and release permissions are implementation decisions still to be defined.
 ### Trust and authority boundaries
@@ -57,10 +60,12 @@ Central controls verify protected-branch review, actual artifact identity, exact
 </details>
 <details>
-<summary>Responsibilities and first workflows for all eleven teams</summary>
+<summary>Candidate contributions and first workflows for all eleven teams</summary>
 ### Department contributions
 The existing team structure provides a starting point. The contributions below are implementation proposals, not settled approval roles or lifecycle gates.
 ### Product / Product Management
 **Contribution and retained human decision** — Own pilot priorities, acceptance criteria and outcome hypotheses; decide product value.
@@ -133,6 +138,30 @@ See the existing [organization structure](/organization/structure/), [department
 </details>
 <details>
 <summary>Strategy rationale and reading map</summary>
 ### Govern skills first, expand authority only with evidence
 Eldorado's target is to make AI-assisted work dependable across the organization: useful guidance, clear human ownership, controlled execution and measurable outcomes. The first investment is a governed skills system, not an autonomous replacement for the organization. A skill packages reusable working knowledge, prerequisites and verification steps. It helps an agent perform a task; it does not grant permission or guarantee that the work is correct.
 The target combines an autonomous agent executing and governing the SDLC with humans directing, guiding and reviewing it. Developer choice on local laptops remains a supporting option: Copilot, Claude Code and Cline can receive reviewed portable skill content through qualified adapters. Repository, CI and release controls support execution; detailed approval policies remain implementation decisions, not new roles established by this strategy.
 The strategy describes the intended operating model and the evidence required to introduce it. It does not establish that these controls are installed, that every client is supported, or that AI has improved Eldorado's performance. The current reference is the documented six-department organization. Rollout begins by validating actual clients, responsibilities, data boundaries and available evidence.
 Success means accepted outcomes with maintained quality, controlled risk and understood total cost. Faster activity without reliable outcomes is not success. More skill invocations, generated lines or tool subscriptions are not productivity targets. People remain accountable for product, customer, financial and personnel decisions.
 ### Navigate the strategy
 - [Roadmap](./roadmap.mdx) — Evidence-gated phases, no promised dates or budgets.
 - [Skills governance](./skills-governance.mdx) — Contract, lifecycle, decision rights, adapters, locking, exceptions.
 - [Skills quality](./skills-quality.mdx) — Risk tiers, independent evaluation, repeated trials, adversarial conformance.
 - [Skills adoption](./skills-adoption.mdx) — Workflow-led coaching, client choice, feedback, manual path.
 - [Measurement](./measurement.mdx) — Full KPI contracts, minimal events, illustrative calculations, privacy and limits.
 - [Skills maintenance](./skills-maintenance.mdx) — Review, drift, deprecation, rollback and incident handling.
 </details>
 **Next:** [Decide who owns and approves each skill →](./skills-governance.mdx)
 </div>
@@ -18,6 +18,7 @@ const config: Config = {
    theme: {customCss: './src/css/custom.css'},
  } satisfies Preset.Options]],
  plugins: [
    ['@docusaurus/plugin-client-redirects', {redirects: [{from: '/strategy/', to: '/strategy/target-operating-model/'}]}],
    ['@docusaurus/plugin-content-docs', {id: 'organization', path: 'docs/organization', routeBasePath: 'organization', sidebarPath: './sidebars/organization.ts'}],
    ['@docusaurus/plugin-content-docs', {id: 'strategy', path: 'docs/strategy', routeBasePath: 'strategy', sidebarPath: './sidebars/strategy.ts'}],
    ['@docusaurus/plugin-content-docs', {id: 'contribution', path: 'docs/contribution', routeBasePath: 'contribution', sidebarPath: './sidebars/contribution.ts'}],
@@ -15,11 +15,12 @@
    "verify:build": "node scripts/verify.mjs --build",
    "diagrams:build": "node scripts/build-diagrams.mjs",
    "test": "node --test scripts/*.test.mjs",
-    "test:browser": "node scripts/browser/strategy-story.cjs && node scripts/browser/inline-controls.cjs",
+    "test:browser": "node scripts/browser/strategy-story.cjs && node scripts/browser/inline-controls.cjs && node scripts/browser/sdlc-cycle.cjs && node scripts/browser/skills-measurement.cjs",
    "kpi:example": "node scripts/ai-kpi.mjs"
  },
  "dependencies": {
    "@docusaurus/core": "3.10.2",
    "@docusaurus/plugin-client-redirects": "3.10.2",
    "@docusaurus/preset-classic": "3.10.2",
    "@docusaurus/theme-common": "3.10.2",
    "@easyops-cn/docusaurus-search-local": "0.55.3",
@@ -14,6 +14,9 @@ importers:
      '@docusaurus/core':
        specifier: 3.10.2
        version: 3.10.2(@mdx-js/react@3.1.1(@types/react@19.1.13)(react@19.1.1))(debug@4.4.3)(postcss@8.5.28)(react-dom@19.1.1(react@19.1.1))(react@19.1.1)(typescript@5.9.3)
      '@docusaurus/plugin-client-redirects':
        specifier: 3.10.2
        version: 3.10.2(@mdx-js/react@3.1.1(@types/react@19.1.13)(react@19.1.1))(postcss@8.5.28)(react-dom@19.1.1(react@19.1.1))(react@19.1.1)(typescript@5.9.3)
      '@docusaurus/preset-classic':
        specifier: 3.10.2
        version: 3.10.2(@algolia/client-search@5.59.0)(@mdx-js/react@3.1.1(@types/react@19.1.13)(react@19.1.1))(@types/react@19.1.13)(postcss@8.5.28)(react-dom@19.1.1(react@19.1.1))(react@19.1.1)(search-insights@2.17.3)(typescript@5.9.3)
@@ -1115,6 +1118,13 @@ packages:
      react: '*'
      react-dom: '*'
  '@docusaurus/plugin-client-redirects@3.10.2':
    resolution: {integrity: sha512-z5I5ttCXw+8y2gHVZvqAMmUw4Rb0ZzKA5eCPk87SfM/jCKOTvE24yDpPAwwipvug96ij7mOwEHXWw8G4LlWdGA==}
    engines: {node: '>=20.0'}
    peerDependencies:
      react: ^18.0.0 || ^19.0.0
      react-dom: ^18.0.0 || ^19.0.0
  '@docusaurus/plugin-content-blog@3.10.2':
    resolution: {integrity: sha512-0cbEnNKf0InmLkhj/+nVRmqEnWEoOE8Mh+2x1qOXI0qYpCnphq4RXknVJ8BvybKRXqYVvbmdMfiJSup+k4tm5w==}
    engines: {node: '>=20.0'}
@@ -6784,6 +6794,47 @@ snapshots:
      - uglify-js
      - webpack-cli
  '@docusaurus/plugin-client-redirects@3.10.2(@mdx-js/react@3.1.1(@types/react@19.1.13)(react@19.1.1))(postcss@8.5.28)(react-dom@19.1.1(react@19.1.1))(react@19.1.1)(typescript@5.9.3)':
    dependencies:
      '@docusaurus/core': 3.10.2(@mdx-js/react@3.1.1(@types/react@19.1.13)(react@19.1.1))(debug@4.4.3)(postcss@8.5.28)(react-dom@19.1.1(react@19.1.1))(react@19.1.1)(typescript@5.9.3)
      '@docusaurus/logger': 3.10.2
      '@docusaurus/utils': 3.10.2(clean-css@5.3.3)(cssnano@6.1.2(postcss@8.5.28))(html-minifier-terser@7.2.0)(postcss@8.5.28)(react-dom@19.1.1(react@19.1.1))(react@19.1.1)
      '@docusaurus/utils-common': 3.10.2(clean-css@5.3.3)(cssnano@6.1.2(postcss@8.5.28))(html-minifier-terser@7.2.0)(postcss@8.5.28)(react-dom@19.1.1(react@19.1.1))(react@19.1.1)
      '@docusaurus/utils-validation': 3.10.2(postcss@8.5.28)(react-dom@19.1.1(react@19.1.1))(react@19.1.1)
      eta: 2.2.0
      fs-extra: 11.4.1
      lodash: 4.18.1
      react: 19.1.1
      react-dom: 19.1.1(react@19.1.1)
      tslib: 2.8.1
    transitivePeerDependencies:
      - '@docusaurus/faster'
      - '@mdx-js/react'
      - '@minify-html/node'
      - '@napi-rs/image'
      - '@parcel/css'
      - '@rspack/core'
      - '@swc/core'
      - '@swc/css'
      - '@swc/html'
      - bufferutil
      - clean-css
      - cssnano
      - csso
      - debug
      - esbuild
      - html-minifier-terser
      - imagemin
      - lightningcss
      - postcss
      - sharp
      - supports-color
      - svgo
      - typescript
      - uglify-js
      - utf-8-validate
      - webpack-cli
  '@docusaurus/plugin-content-blog@3.10.2(@docusaurus/plugin-content-docs@3.10.2(@mdx-js/react@3.1.1(@types/react@19.1.13)(react@19.1.1))(postcss@8.5.28)(react-dom@19.1.1(react@19.1.1))(react@19.1.1)(typescript@5.9.3))(@mdx-js/react@3.1.1(@types/react@19.1.13)(react@19.1.1))(postcss@8.5.28)(react-dom@19.1.1(react@19.1.1))(react@19.1.1)(typescript@5.9.3)':
    dependencies:
      '@docusaurus/core': 3.10.2(@mdx-js/react@3.1.1(@types/react@19.1.13)(react@19.1.1))(debug@4.4.3)(postcss@8.5.28)(react-dom@19.1.1(react@19.1.1))(react@19.1.1)(typescript@5.9.3)
@@ -28,11 +28,6 @@
    ]
  },
  "pages": [
    {
      "id": "index",
      "title": "AI adoption strategy",
      "purpose": "Govern skills first; outcomes before activity; explicit target status."
    },
    {
      "id": "target-operating-model",
      "title": "Target operating model",
@@ -55,8 +50,8 @@
    },
    {
      "id": "measurement",
-      "title": "Measurement",
+      "title": "Skills Measurement",
-      "purpose": "Full KPI contracts, minimal events, illustrative calculations, privacy and limits."
+      "purpose": "Which KPIs to collect during skills adoption, how to calculate them and when to report."
    },
    {
      "id": "skills-maintenance",
@@ -25,6 +25,21 @@ const fs = require('node:fs');
        const box = await frame.locator(`#${id}`).boundingBox();
        assert(box && box.width >= 40 && box.height >= 44, id);
      }
      const toolbar = await frame.locator('.toolbar').boundingBox();
      const exportButton = await frame.locator('#btn-export').boundingBox();
      const diagram = await frame.locator('.diagram-container > svg').boundingBox();
      const rightGap = toolbar.x + toolbar.width - exportButton.x - exportButton.width;
      assert(Math.abs(rightGap - 8) < 1, `toolbar right padding at ${width}: ${rightGap}`);
      assert(diagram.y >= toolbar.y + toolbar.height, 'toolbar must stay above nodes');
      const checkMenu = async selector => {
        await frame.locator(selector).evaluate(n => Promise.all(n.getAnimations().map(a => a.finished)));
        const menu = await frame.locator(selector).boundingBox();
        const viewport = await iframe.boundingBox();
        const currentToolbar = await frame.locator('.toolbar').boundingBox();
        assert(menu && menu.x >= viewport.x - 1 && menu.y >= currentToolbar.y + currentToolbar.height - 1, selector);
        assert(menu.x + menu.width <= viewport.x + viewport.width + 1, `${selector} clips horizontally`);
        assert(menu.y + menu.height <= viewport.y + viewport.height + 1, `${selector} clips vertically`);
      };
      const light = await frame.locator('body').evaluate(n => getComputedStyle(n).backgroundColor);
      await frame.locator('#btn-theme').click(); await theme('dark');
      assert.notEqual(await frame.locator('body').evaluate(n => getComputedStyle(n).backgroundColor), light);
@@ -42,6 +57,7 @@ const fs = require('node:fs');
      await portalTheme('light'); await theme('light');
      for (const preset of ['signal-flow', 'blueprint', 'editorial', 'classic']) {
        await frame.locator('#btn-preset').click();
        await checkMenu('.preset-menu');
        await frame.locator(`[data-preset-value="${preset}"]`).click();
        assert.equal(await frame.locator('html').getAttribute('data-preset'), preset);
      }
@@ -53,6 +69,7 @@ const fs = require('node:fs');
      const exports = [];
      for (const format of ['png', 'jpeg', 'webp', 'svg', 'svg-light', 'svg-dark', 'webm']) {
        await frame.locator('#btn-export').click();
        await checkMenu('.export-menu');
        if (format === 'webm' && await frame.locator('[data-format="webm"]').isDisabled()) {
          exports.push({format, unsupported: true});
          await frame.locator('#btn-export').click();
@@ -71,7 +88,10 @@ const fs = require('node:fs');
        if (format.startsWith('svg')) {
          const valid = await page.evaluate(xml => {
            const doc = new DOMParser().parseFromString(xml, 'image/svg+xml');
-            return !doc.querySelector('parsererror') && doc.documentElement.localName === 'svg' && doc.querySelectorAll('text').length > 0;
+            if (doc.querySelector('[data-node-id="humans"], [data-edge-id="human-direction"]')) return false;
            if (![...doc.querySelectorAll('[data-node-id="agent"] text')].some(n => n.textContent === 'Human Controlled')) return false;
            if (doc.querySelectorAll('[data-node-id]').length !== 17 || doc.querySelectorAll('path[data-edge-id]').length !== 24) return false;
            return !doc.querySelector('parsererror') && doc.documentElement.localName === 'svg' && doc.querySelectorAll('text').length > 0 && doc.querySelector('#eldorado-cycle-export')?.textContent.includes('stroke-dasharray: 3 9');
          }, bytes.toString());
          assert(valid, format);
        } else {
@@ -100,15 +120,22 @@ const fs = require('node:fs');
      await close.focus(); await page.keyboard.press('Escape');
      assert(await opener.evaluate(n => n === document.activeElement));
      await iframe.scrollIntoViewIfNeeded();
-      await page.evaluate(() => scrollTo(0, 0));
+      await page.evaluate(() => scrollTo({top: 0, behavior: 'instant'}));
      await page.evaluate(() => new Promise(resolve => requestAnimationFrame(() => requestAnimationFrame(resolve))));
      await page.screenshot({path:`${output}/${width}.png`, fullPage:true});
      await frame.locator('#btn-theme').click(); await theme('dark');
-      await page.locator('.strategy-content a[href$="/strategy/target-operating-model/"]').filter({visible:true}).click();
+      // Exercise keyboard navigation without racing the portal's smooth scroll.
      await page.locator('.strategy-content a[href$="/strategy/skills-governance/"]').filter({visible:true}).focus();
      await page.keyboard.press('Enter');
      await page.waitForURL('**/strategy/skills-governance/');
      await page.goBack({waitUntil:'domcontentloaded'});
      await page.waitForURL('**/strategy/target-operating-model/');
      await page.getByRole('heading', {name:'Target operating model', exact:true, level:1}).waitFor();
      const next = page.locator(`.archify-panel--story[data-diagram="strategy-target"] > .story-${width > 700 ? 'desktop' : 'mobile'}`);
      await next.scrollIntoViewIfNeeded();
      await (await (await next.elementHandle()).contentFrame()).waitForFunction(() => document.documentElement.dataset.theme === 'dark');
      assert(!await page.evaluate(() => document.documentElement.scrollWidth > innerWidth));
-      results.push({width, themeSync:true, presets:4, exports, fullscreen:true, routeTheme:true});
+      results.push({width, toolbar, rightGap, dropdownsUnclipped:true, themeSync:true, presets:4, exports, fullscreen:true, routeTheme:true});
      fs.writeFileSync(`${output}/results.json`, JSON.stringify(results,null,2));
      await context.close();
    }
@@ -0,0 +1,90 @@
 const {chromium} = require('playwright');
 const assert = require('node:assert/strict');
 const fs = require('node:fs');
 (async () => {
  const base = process.argv[2] || 'http://127.0.0.1:4173';
  const output = `.cache/cycle-${process.argv[3] || 'local'}`;
  fs.mkdirSync(output, {recursive:true});
  const browser = await chromium.launch({headless:true, executablePath:process.env.PLAYWRIGHT_CHROMIUM_EXECUTABLE, args:['--no-sandbox']});
  const results = [];
  try {
    for (const width of [1440, 390, 320]) {
      const context = await browser.newContext({viewport:{width,height:1000}, colorScheme:'light'});
      const page = await context.newPage();
      await page.goto(`${base}/eldorado-v1/documentation/`, {waitUntil:'networkidle'});
      if (width < 700) await page.getByRole('button', {name:'Toggle navigation bar'}).click();
      const nav = page.getByRole('link', {name:'Strategy', exact:true}).filter({visible:true});
      assert((await nav.getAttribute('href')).endsWith('/strategy/target-operating-model/'));
      await nav.click();
      await page.waitForURL('**/strategy/target-operating-model/');
      for (const route of ['target-operating-model/', '']) {
        await page.goto(`${base}/eldorado-v1/documentation/strategy/${route}`, {waitUntil:'networkidle'});
        await page.waitForURL('**/strategy/target-operating-model/');
        if (width < 700) await page.getByRole('button', {name:'Toggle navigation bar'}).click();
        const menu = page.locator('.theme-doc-sidebar-menu .menu__link').filter({visible:true});
        await menu.first().waitFor({state:'visible'});
        assert.deepEqual(await menu.allTextContents(), ['Target', 'Roadmap', 'Skills governance', 'Skills quality', 'Skills adoption', 'Skills Measurement', 'Skills maintenance']);
        assert((await menu.nth(0).getAttribute('href')).endsWith('/strategy/target-operating-model/'));
        assert((await menu.nth(1).getAttribute('href')).endsWith('/strategy/roadmap/'));
        assert.equal(await menu.filter({hasText:/^Target$/}).count(), 1);
        await page.screenshot({path:`${output}/${width}-${route ? 'target' : 'landing'}-navigation.png`,fullPage:true,animations:'disabled'});
        await menu.nth(1).click();
        await page.waitForURL('**/strategy/roadmap/');
        await page.getByRole('heading', {level:1,name:'Roadmap',exact:true}).waitFor();
        await page.goBack({waitUntil:'domcontentloaded'});
        await page.waitForURL('**/strategy/target-operating-model/');
        const iframe = page.locator(`.archify-panel--story > .story-${width > 700 ? 'desktop' : 'mobile'}`);
        await iframe.scrollIntoViewIfNeeded();
        const frame = await (await iframe.elementHandle()).contentFrame();
        await frame.locator('svg[role="img"]').waitFor();
        const metrics = await frame.evaluate(() => {
          const svg = document.querySelector('svg[role="img"]');
          const labels = [...svg.querySelectorAll('text[data-node-label]')].map(n => ({text:n.textContent,px:parseFloat(getComputedStyle(n).fontSize)*n.getScreenCTM().a}));
          const edges = [...svg.querySelectorAll('path[data-edge-id^="stage-"]')].map(n => [n.dataset.edgeFrom,n.dataset.edgeTo,n.getAttribute('marker-end')]);
          const overflow = [...svg.querySelectorAll('[data-node-id]')].flatMap(n => {
            const box = n.querySelector('rect').getBoundingClientRect();
            return [...n.querySelectorAll('text')].filter(t => {const b=t.getBoundingClientRect();return b.left<box.left||b.right>box.right||b.top<box.top||b.bottom>box.bottom;}).map(t=>t.textContent);
          });
          const relationships = [...svg.querySelectorAll('path[data-edge-id]')].map(n => ({id:n.dataset.edgeId,from:n.dataset.edgeFrom,to:n.dataset.edgeTo,dash:getComputedStyle(n).strokeDasharray,arrow:n.getAttribute('marker-end')}));
          const nodes = [...svg.querySelectorAll('[data-node-id]')].map(n => {const r=n.querySelector('rect').getBoundingClientRect();return {id:n.dataset.nodeId,x:r.x+r.width/2,y:r.y+r.height/2};});
          return {labels,edges,overflow,relationships,nodes};
        });
        const ids = ['scope','architecture','ui-ux','planning','delivery','uat','security','releases'];
        assert.deepEqual(metrics.edges.map(e=>e.slice(0,2)), ids.map((id,i)=>[id,ids[(i+1)%ids.length]]));
        assert(metrics.edges.every(e=>e[2]?.startsWith('url(')), 'all stages must have arrowheads');
        assert.equal(metrics.nodes.length, 17);
        assert(!metrics.nodes.some(n=>n.id==='humans'));
        assert.equal(await frame.locator('[data-node-id="agent"] text').allTextContents().then(texts=>texts.includes('Human Controlled')), true);
        assert.equal(metrics.relationships.length, 24);
        const governs = metrics.relationships.filter(e=>e.id.startsWith('govern-'));
        const contributions = metrics.relationships.filter(e=>e.id.startsWith('contribute-'));
        assert.deepEqual(governs.map(e=>[e.from,e.to]), ids.map(id=>['agent',id]));
        assert.deepEqual(contributions.map(e=>[e.from,e.to]), ids.map(id=>[`code-${id}`,id]));
        assert(governs.every(e=>e.dash==='3px, 9px' && e.arrow?.startsWith('url(')));
        assert(contributions.every(e=>e.dash==='12px, 8px' && e.arrow?.startsWith('url(')));
        assert(!metrics.relationships.some(e=>e.id==='human-direction' || e.from==='humans' || e.to==='humans'));
        if (width > 700) {
          const agent = metrics.nodes.find(n=>n.id==='agent');
          const phases = metrics.nodes.filter(n=>ids.includes(n.id));
          for (const axis of ['x','y']) assert(Math.abs(agent[axis]-(Math.min(...phases.map(n=>n[axis]))+Math.max(...phases.map(n=>n[axis])))/2)<1);
          for (const id of ids) {
            const phase=phases.find(n=>n.id===id), code=metrics.nodes.find(n=>n.id===`code-${id}`);
            assert(Math.hypot(code.x-agent.x,code.y-agent.y)>Math.hypot(phase.x-agent.x,phase.y-agent.y));
          }
        }
        assert(metrics.labels.every(n=>n.px>=14), JSON.stringify(metrics.labels));
        assert.deepEqual(metrics.overflow, []);
        assert(!await page.evaluate(()=>document.documentElement.scrollWidth>innerWidth));
        await page.evaluate(()=>scrollTo({top:0,behavior:'instant'}));
        await page.waitForFunction(()=>scrollY===0);
        await page.evaluate(()=>new Promise(resolve=>requestAnimationFrame(()=>requestAnimationFrame(resolve))));
        await page.screenshot({path:`${output}/${width}-${route ? 'target' : 'landing'}.png`,fullPage:true});
        results.push({width,route,...metrics});
        fs.writeFileSync(`${output}/results.json`,JSON.stringify(results,null,2));
      }
      await context.close();
    }
    assert.equal(results.length,6);
    console.log('PASS 6 cycle routes/viewports: navigation, sidebar order, exact directed loop, arrowheads, >=14px labels, text bounds and no overflow');
  } finally {await browser.close();}
 })().catch(error=>{console.error(error);process.exitCode=1;});
@@ -0,0 +1,85 @@
 const {chromium} = require('playwright');
 const assert = require('node:assert/strict');
 const fs = require('node:fs');
 (async () => {
  const base = process.argv[2] || 'http://127.0.0.1:4173';
  const output = `.cache/measurement-${process.argv[3] || 'local'}`;
  fs.mkdirSync(output, {recursive:true});
  const browser = await chromium.launch({headless:true, executablePath:process.env.PLAYWRIGHT_CHROMIUM_EXECUTABLE, args:['--no-sandbox']});
  const results = [];
  try {
    for (const width of [1440, 390, 320]) {
      const context = await browser.newContext({viewport:{width,height:1000}, colorScheme:'light'});
      const page = await context.newPage();
      const errors = [];
      page.on('pageerror', error => errors.push(error.message));
      const response = await page.goto(`${base}/eldorado-v1/documentation/strategy/measurement/`, {waitUntil:'networkidle'});
      assert.equal(response.status(), 200);
      assert.equal(await page.locator('h1').innerText(), 'Skills Measurement');
      assert((await page.title()).startsWith('Skills Measurement |'));
      assert.equal(await page.locator('.strategy-content details').count(), 0);
      assert.equal(await page.locator('.strategy-content h2').count(), 4);
      if (width < 700) await page.getByRole('button', {name:'Toggle navigation bar'}).click();
      const menu = page.locator('.theme-doc-sidebar-menu .menu__link').filter({visible:true});
      await menu.first().waitFor({state:'visible'});
      assert.deepEqual(await menu.allTextContents(), ['Target','Roadmap','Skills governance','Skills quality','Skills adoption','Skills Measurement','Skills maintenance']);
      if (width < 700) await page.getByRole('button', {name:'Close navigation bar'}).click();
      const panel = page.locator('.archify-panel--story');
      const iframe = panel.locator(width > 700 ? '.story-desktop' : '.story-mobile');
      await iframe.scrollIntoViewIfNeeded();
      const frame = await (await iframe.elementHandle()).contentFrame();
      await frame.locator('svg[role="img"]').waitFor();
      const geometry = await frame.locator('svg[role="img"]').evaluate(svg => {
        const nodes = [...svg.querySelectorAll('[data-node-id]')].map(n => n.getBoundingClientRect().toJSON());
        return {nodes, width:innerWidth, height:innerHeight, minLabel:Math.min(...[...svg.querySelectorAll('text[data-node-label]')].map(n => parseFloat(getComputedStyle(n).fontSize)*n.getScreenCTM().a))};
      });
      assert.equal(geometry.nodes.length, 6);
      for (const n of geometry.nodes) assert(n.x >= 0 && n.y >= 0 && n.right <= geometry.width + 1 && n.bottom <= geometry.height + 1);
      assert(geometry.minLabel >= 10, `readable node labels: ${geometry.minLabel}`);
      await frame.locator('#btn-theme').click();
      await page.waitForFunction(() => document.documentElement.dataset.theme === 'dark');
      await frame.locator('#btn-theme').click();
      await page.waitForFunction(() => document.documentElement.dataset.theme === 'light');
      for (const preset of ['blueprint','classic']) {
        await frame.locator('#btn-preset').click();
        await frame.locator(`[data-preset-value="${preset}"]`).click();
        assert.equal(await frame.locator('html').getAttribute('data-preset'), preset);
      }
      await frame.locator('#btn-export').click();
      const [download] = await Promise.all([page.waitForEvent('download'), frame.locator('[data-format="svg"]').click()]);
      assert.equal(await download.failure(), null);
      const path = `${output}/${width}.svg`; await download.saveAs(path);
      const svg = fs.readFileSync(path, 'utf8');
      assert(svg.includes('Establish baseline') && svg.includes('Ongoing monitoring'));
      const opener = panel.getByRole('button', {name:/^Expand /});
      await opener.click();
      const close = panel.getByRole('button', {name:'Close full-screen diagram', exact:true});
      await close.waitFor({state:'visible'}); assert.equal(await close.innerText(), '×');
      await close.click(); assert(await opener.evaluate(n => n === document.activeElement));
      await opener.click(); await close.waitFor({state:'visible'}); await close.focus(); await page.keyboard.press('Escape');
      await page.waitForFunction(() => !document.querySelector('.archify-panel--story dialog').open);
      assert(await opener.evaluate(n => n === document.activeElement));
      assert(!await page.evaluate(() => document.documentElement.scrollWidth > innerWidth));
      await page.evaluate(() => scrollTo({top:0, behavior:'instant'}));
      await page.screenshot({path:`${output}/${width}-top.png`});
      await page.locator('.strategy-content h2').nth(1).scrollIntoViewIfNeeded();
      await page.screenshot({path:`${output}/${width}-pilot.png`});
      await page.screenshot({path:`${output}/${width}-full.png`, fullPage:true});
      const links = await page.locator('.strategy-content a[href*="measurement-contracts"]').evaluateAll(nodes => nodes.map(n => n.href));
      for (const href of [...new Set(links)]) {
        const navigation = await page.goto(href, {waitUntil:'networkidle'});
        if (navigation) assert.equal(navigation.status(), 200);
        assert.equal(page.url(), href);
        if (new URL(href).hash) await page.locator(new URL(href).hash).waitFor({state:'visible'});
      }
      assert.equal(await page.locator('.strategy-content h3').count(), 19);
      assert((await page.locator('.strategy-content').innerText()).includes('9 prototype calculations and 7 unimplemented metrics'));
      assert(!await page.evaluate(() => document.documentElement.scrollWidth > innerWidth));
      assert.deepEqual(errors, []);
      results.push({width, geometry, links:links.length, title:true, sidebar:true, controls:true, export:true, reference:true});
      fs.writeFileSync(`${output}/results.json`, JSON.stringify(results,null,2));
      await context.close();
    }
    console.log(`PASS ${results.length} Skills Measurement viewports: title/sidebar, open plan, diagram geometry/controls/export, all reference links and no page errors`);
  } finally {await browser.close();}
 })().catch(error => {console.error(error); process.exitCode = 1;});
@@ -19,23 +19,28 @@ const fs=require('node:fs');
    const nodes=[...svg.querySelectorAll('[data-node-id]')].map(n=>{const r=n.getBoundingClientRect();return {id:n.id,x:r.x,y:r.y,right:r.right,bottom:r.bottom};});
    return {width:innerWidth,height:innerHeight,nodes,svg:svg.getBoundingClientRect().toJSON(),minLabel:Math.min(...[...svg.querySelectorAll('text[data-node-label]')].map(n=>parseFloat(getComputedStyle(n).fontSize)*n.getScreenCTM().a))};
   });
-   assert(geometry.nodes.length>=5 && geometry.nodes.length<=7,id);
+   assert(geometry.nodes.length>=5 && geometry.nodes.length<=(['index','target-operating-model'].includes(id)?18:7),id);
   for(const n of geometry.nodes) assert(n.x>=0 && n.y>=0 && n.right<=geometry.width+1 && n.bottom<=geometry.height+1,`${id} ${width} clipped ${JSON.stringify(n)} / ${geometry.height}`);
   const data=await page.locator('.strategy-content').evaluate(el=>({words:el.innerText.trim().split(/\s+/).length,overflow:document.documentElement.scrollWidth>innerWidth,openDetails:el.querySelectorAll('details[open]').length}));
-   assert(data.words<=200,id);assert(!data.overflow,id);assert.equal(data.openDetails,0,id);
+   assert(data.words<=(id==='measurement'?1100:200),id);assert(!data.overflow,id);assert.equal(data.openDetails,0,id);
-   await page.evaluate(()=>scrollTo(0,0));await page.screenshot({path:`.cache/story-${label}/${width}-${id}.png`,fullPage:true});
+   await page.evaluate(()=>scrollTo({top:0,behavior:'instant'}));await page.screenshot({path:`.cache/story-${label}/${width}-${id}.png`,fullPage:true});
   const opener=panel.getByRole('button',{name:/^Expand /});await opener.click();
   const close=panel.getByRole('button',{name:'Close full-screen diagram',exact:true});await close.waitFor({state:'visible'});
   assert.equal(await close.innerText(),'×');assert(await close.evaluate(n=>n===document.activeElement));
   const box=await close.boundingBox();assert.equal(box.width,44);assert.equal(box.height,44);
   await close.click();assert(await opener.evaluate(n=>n===document.activeElement));
   await opener.click();await close.waitFor({state:'visible'});await page.keyboard.press('Escape');await page.waitForFunction(()=>!document.querySelector('.archify-panel--story dialog').open);assert(await opener.evaluate(n=>n===document.activeElement));
-   const summary=page.locator('.strategy-content > details > summary').first();await summary.click();assert(await summary.evaluate(n=>n.parentElement.open));await summary.click();
+   if(id !== 'measurement') {
-   results.push({id,width,...data,geometry,modalClick:true,modalEscape:true,focusRestored:true,details:true});fs.writeFileSync(`.cache/story-${label}/results.json`,JSON.stringify(results,null,2));
+   const summary=page.locator('.strategy-content > details > summary').first();
   await summary.focus();await page.keyboard.press('Enter');
   await page.waitForFunction(()=>document.querySelector('.strategy-content > details').open);
   await page.keyboard.press('Enter');
   }
   results.push({id,width,...data,geometry,modalClick:true,modalEscape:true,focusRestored:true,details:id!=='measurement'});fs.writeFileSync(`.cache/story-${label}/results.json`,JSON.stringify(results,null,2));
  }
-  await page.goto(`${base}/eldorado-v1/documentation/strategy/measurement/#validated-task-success`,{waitUntil:'networkidle'});
+  await page.goto(`${base}/eldorado-v1/documentation/strategy/reference/measurement-contracts/#validated-task-success`,{waitUntil:'networkidle'});
  const heading=page.locator('#validated-task-success');await heading.waitFor({state:'visible'});
-  assert(await heading.evaluate(n=>n.closest('details').open),'hash should open metric reference');
+  assert.equal(await page.locator('.strategy-content details').count(),0,'reference is directly readable');
  await context.close();
 }
 console.log(`PASS ${results.length} desktop/mobile cases, clipping, default word count, details, modal click/Escape/focus and KPI hash navigation`);
@@ -4,6 +4,7 @@ import {execFileSync} from 'node:child_process';
 const cli = 'vendor/archify/bin/archify.mjs';
 const types = new Set(['architecture','workflow','sequence','dataflow','lifecycle']);
 const inlineStyles = readFileSync(new URL('./inline-viewer.css', import.meta.url), 'utf8');
 const cycleStyles = readFileSync(new URL('./cycle-viewer.css', import.meta.url), 'utf8');
 const themeBridge = readFileSync(new URL('./viewer-theme-bridge.js', import.meta.url), 'utf8');
 // This directory is generated-only and ignored by Git. Remove stale artifacts.
 rmSync('static/diagrams', {recursive: true, force: true});
@@ -36,7 +37,16 @@ for (const file of readdirSync('diagrams').sort()) {
  if (start < 0 || end < start || preset.split(guard).length !== 3) throw new Error('Archify preset integration drift');
  viewer = viewer.slice(0, start) + preset.replaceAll(guard, '') + viewer.slice(end);
  if (!viewer.includes('</head>') || !viewer.includes('</body>')) throw new Error(`Invalid viewer: ${output}`);
-  writeFileSync(output, viewer.replace('</head>', `<style id="eldorado-inline-controls">${inlineStyles}</style></head>`)
+  const storyStyles = ['strategy-target', 'strategy-target-mobile'].includes(name)
    ? `${name === 'strategy-target' ? ':root {--cycle-label: 27px; --cycle-detail: 21px;}' : ':root {--cycle-label: 38px; --cycle-detail: 28px;}'}${cycleStyles}` : '';
  if (storyStyles) {
    // Keep typography and relationship styles inside exported standalone SVGs.
    const diagramRoot = /<svg\b[^>]*role="img"[^>]*>/g;
    if ([...viewer.matchAll(diagramRoot)].length !== 1) throw new Error('Cycle SVG integration drift');
    const exportStyles = storyStyles.replaceAll(':root', 'svg[role="img"]').replaceAll('.diagram-container > svg', 'svg[role="img"]');
    viewer = viewer.replace(diagramRoot, root => `${root}<style id="eldorado-cycle-export">${exportStyles}</style>`);
  }
  writeFileSync(output, viewer.replace('</head>', `<style id="eldorado-inline-controls">${inlineStyles}</style><style id="eldorado-cycle-labels">${storyStyles}</style></head>`)
    .replace('</body>', `<script>${themeBridge}</script></body>`));
  renameSync(`static/diagrams/${name}/index.delivery.json`, `.cache/archify/${name}.delivery.json`);
 }
@@ -0,0 +1,19 @@
 /* Readable lifecycle story at inline scale, isolated from other diagrams. */
 .diagram-container > svg text {
  font-family: Arial, sans-serif;
  letter-spacing: 0;
  font-size: 30px;
 }
 .diagram-container > svg text[data-node-label] {
  font-size: var(--cycle-label);
  transform: translateY(-6px);
 }
 .diagram-container > svg text[data-detail="context"] {
  font-size: var(--cycle-detail);
  transform: translateY(16px);
 }
 /* Edge semantics stay in SVG metadata; one page legend avoids repeated labels. */
 .diagram-container > svg g[data-edge-id] {display: none;}
 .diagram-container > svg path[data-edge-id^="stage-"] {stroke-width: 2.5;}
 .diagram-container > svg path[data-edge-id^="govern-"] {stroke-dasharray: 3 9; stroke-width: 3;}
 .diagram-container > svg path[data-edge-id^="contribute-"] {stroke-dasharray: 12 8; stroke-width: 3;}
@@ -3,9 +3,9 @@ html[data-embed="true"] .toolbar {
  display: flex !important;
  position: relative;
  inset: auto;
-  justify-content: center;
+  justify-content: flex-end;
  height: 52px;
-  padding: 4px;
+  padding: 4px 8px;
  box-sizing: border-box;
  background: var(--bg);
 }
@@ -16,17 +16,18 @@ html[data-embed="true"] .diagram-container > svg {
 }
 html[data-embed="true"] .toolbar .preset-menu,
 html[data-embed="true"] .toolbar .export-menu {
-  position: absolute;
+  position: fixed;
-  top: 100%;
+  top: 52px;
-  left: 0;
+  left: auto;
-  right: auto;
+  right: 0;
  width: min(320px, 100vw);
  max-width: 100vw;
  max-height: calc(100vh - 52px);
  overflow-y: auto;
  box-sizing: border-box;
 }
 /* Keep the preset menu in the iframe viewport without moving its capsule. */
 html[data-embed="true"] .toolbar .export-wrap { position: static; }
-html[data-embed="true"] .toolbar .export-menu { left: auto; right: 0; }
+html[data-embed="true"] .toolbar .export-menu { position: absolute; top: 100%; }
 html[data-embed="true"] .toolbar .preset-menu button,
 html[data-embed="true"] .toolbar .export-menu button { min-height: 44px; }
@@ -0,0 +1,61 @@
 import {test} from 'node:test';
 import assert from 'node:assert/strict';
 import {readFileSync} from 'node:fs';
 const read = p => readFileSync(p, 'utf8');
 const stages = ['Scope', 'Architecture', 'UI/UX', 'Planning', 'Delivery', 'UAT', 'Security', 'Releases'];
 test('Strategy opens at the canonical autonomous lifecycle', () => {
  assert.match(read('sidebars/strategy.ts'), /strategySidebar: \['target-operating-model', 'roadmap'/);
  assert.match(read('docs/strategy/target-operating-model.mdx'), /^sidebar_label: Target$/m);
  for (const id of ['target-operating-model']) {
    const page = read(`docs/strategy/${id}.mdx`);
    const primary = page.split('<details>')[0];
    assert(primary.includes('name="strategy-target"'));
    assert(primary.includes('autonomous agent'));
    assert(primary.includes('direct, guide and review'));
    assert(primary.includes('Scope → Architecture → UI/UX → Planning → Delivery → UAT → Security → Releases → Scope'));
    assert(primary.includes('human-controlled code agent'));
    assert(primary.includes('TO-BE'));
    assert(!/only routes|People authorize|independent gates protect release/.test(primary));
  }
 });
 test('both target diagrams have eight stages, eight governed phases and eight human-controlled contributors', () => {
  for (const suffix of ['', '-mobile']) {
    const d = JSON.parse(read(`diagrams/strategy-target${suffix}.architecture.json`));
    const ids = stages.map(label => d.components.find(n => n.label === label)?.id);
    assert(ids.every(Boolean));
    assert.equal(new Set(ids).size, 8);
    const flow = d.connections.filter(e => ids.includes(e.from) && ids.includes(e.to));
    assert.deepEqual(flow.map(e => [e.from, e.to]), ids.map((id, i) => [id, ids[(i + 1) % ids.length]]));
    assert.match(flow.at(-1).label, /Repeat/);
    assert(d.components.some(n => n.id === 'agent' && n.label === 'Autonomous agent' && n.sublabel === 'Human Controlled'));
    assert(!d.components.some(n => n.id === 'humans' || n.label === 'Humans'));
    assert(!d.connections.some(e => e.id === 'human-direction' || e.from === 'humans' || e.to === 'humans'));
    assert.equal(d.connections.length, 24);
    assert.deepEqual(d.connections.filter(e => e.from === 'agent').map(e => e.to), ids);
    const contributors = d.components.filter(n => n.id.startsWith('code-'));
    assert.equal(contributors.length, 8);
    assert(contributors.every(n => n.label === 'Code agent' && n.sublabel === 'Human-controlled'));
    for (const id of ids) {
      assert(d.connections.some(e => e.from === `code-${id}` && e.to === id && e.variant === 'dashed'));
      assert(d.connections.some(e => e.from === 'agent' && e.to === id && e.variant === 'security'));
    }
    assert.equal(d.components.length, 17);
    if (!suffix) {
      const center = n => n.pos.map((v, i) => v + n.size[i] / 2);
      const agent = center(d.components.find(n => n.id === 'agent'));
      const phaseCenters = ids.map(id => center(d.components.find(n => n.id === id)));
      for (const axis of [0, 1]) {
        const values = phaseCenters.map(c => c[axis]);
        assert.equal(agent[axis], (Math.min(...values) + Math.max(...values)) / 2);
      }
      for (const code of contributors) {
        const phase = d.components.find(n => n.id === code.id.slice(5));
        const distance = n => Math.hypot(...center(n).map((v, i) => v - agent[i]));
        assert(distance(code) > distance(phase), `${code.id} is exterior`);
      }
    }
  }
 });
@@ -0,0 +1,42 @@
 import test from 'node:test';
 import assert from 'node:assert/strict';
 import {readFileSync} from 'node:fs';
 import {calculate} from './ai-kpi.mjs';
 const read = path => readFileSync(path, 'utf8');
 test('Skills Measurement is an open adoption collection plan, not a hidden catalogue', () => {
  const page = read('docs/strategy/measurement.mdx');
  assert.match(page, /title: Skills Measurement\nsidebar_label: Skills Measurement/);
  for (const heading of ['Establish the baseline', 'Measure the pilot', 'Extend collection as adoption grows', 'Keep the collection running']) assert(page.includes(heading), heading);
  assert(!page.includes('<details>'));
  assert(page.includes('./reference/measurement-contracts.mdx'));
  assert(!page.includes('Is AI-assisted work worth continuing?'));
  for (const word of ['Recommended', 'not installed', 'unknown', 'review', 'CI', 'revision', 'Monthly', 'Weekly']) assert(page.includes(word), word);
 });
 test('all sixteen metric definitions remain in the reference with explicit implementation limits', () => {
  const page = read('docs/strategy/reference/measurement-contracts.mdx');
  const model = JSON.parse(read('research/ai-adoption/strategy-model.json'));
  for (const k of model.kpis) {
    const section = page.split(`### ${k.name}\n`)[1]?.split('\n### ')[0];
    assert(section, k.name);
    for (const field of ['Numerator','Denominator','Expression','Aggregation','Unit','Eligible population','Exclusions','Events and source','Owner / cadence','Missing evidence','Gaming and limits']) assert(section.includes(`| ${field} |`));
  }
  for (const text of ['9 prototype calculations', '7 unimplemented metrics', 'no network collectors', '30 days', '12 months', 'below five people', 'Missing telemetry is unknown']) assert(page.includes(text), text);
 });
 test('published pilot arithmetic matches the executable fixture, including partial cost', () => {
  const page = read('docs/strategy/measurement.mdx');
  const fixture = JSON.parse(read('scripts/fixtures/ai-kpi-synthetic.json'));
  const report = calculate(fixture);
  const success = report.kpis['validated-task-success'];
  assert(page.includes(`${success.numerator} / ${success.denominator} × 100 = ${success.value_or_null}%`));
  assert(page.includes(`${success.known_outcome_coverage}%`));
  const cost = report.kpis['cost-per-accepted-outcome'];
  assert.equal(cost.partial_estimate, true);
  assert(page.includes(`EUR ${cost.numerator} / ${cost.denominator} = EUR ${cost.value_or_null}`));
  assert(page.includes('partial estimate'));
  const experience = report.kpis['human-experience'];
  assert(page.includes(`${experience.numerator} / ${experience.denominator} = ${experience.value_or_null}`));
  assert(page.includes(`${experience.denominator} / ${fixture.survey.invited} × 100 = ${experience.coverage.response_percent}%`));
 });
@@ -0,0 +1,19 @@
 import {test} from 'node:test';
 import assert from 'node:assert/strict';
 import {readFileSync, existsSync} from 'node:fs';
 const read = p => readFileSync(p, 'utf8');
 test('Strategy has one canonical Target, then Roadmap, without a landing menu duplicate', () => {
  const sidebar = read('sidebars/strategy.ts').split('strategySidebar: [')[1].split(']')[0];
  const ids = [...sidebar.matchAll(/'([^']+)'/g)].map(match => match[1]);
  assert.deepEqual(ids, ['target-operating-model', 'roadmap', 'skills-governance', 'skills-quality', 'skills-adoption', 'measurement', 'skills-maintenance']);
  assert.equal(ids.filter(id => id === 'target-operating-model').length, 1);
  assert.match(read('docs/strategy/target-operating-model.mdx'), /^sidebar_label: Target$/m);
  assert.match(read('docusaurus.config.ts'), /type: 'docSidebar', docsPluginId: 'strategy', sidebarId: 'strategySidebar', label: 'Strategy'/);
 });
 test('old Strategy entry redirects to Target instead of maintaining a second document', () => {
  assert(!existsSync('docs/strategy/index.mdx'));
  assert.match(read('docusaurus.config.ts'), /'@docusaurus\/plugin-client-redirects'/);
  assert.match(read('docusaurus.config.ts'), /from: '\/strategy\/', to: '\/strategy\/target-operating-model\/'/);
 });
@@ -10,15 +10,15 @@ test('every strategy page leads with a compact diagram story and closes its refe
    const narrative = page.split('<div className="strategy-content">')[1].replace(/<details>[\s\S]*?<\/details>/g, '');
    assert.equal((narrative.match(/<ArchifyDiagram /g) || []).length, 1, id);
    assert(narrative.includes(' story />'), id);
-    assert(page.indexOf(' story />') < page.indexOf('<details>'), id);
+    assert(!page.includes('<details>') || page.indexOf(' story />') < page.indexOf('<details>'), id);
    assert(!/<details\s+open/.test(page), id);
    assert.equal((page.match(/<details>/g) || []).length, (page.match(/<\/details>/g) || []).length, id);
-    assert.equal((narrative.match(/^## /gm) || []).length, 1, id);
+    assert.equal((narrative.match(/^## /gm) || []).length, id === 'measurement' ? 4 : 1, id);
    assert((narrative.match(/^- /gm) || []).length <= 3, id);
    assert(narrative.includes('**Next:**'), id);
    assert(!narrative.includes('| ---'), id);
    const words = narrative.replace(/<[^>]*>/g, '').trim().split(/\s+/).length;
-    assert(words <= 200, `${id}: ${words} default narrative words`);
+    assert(words <= (id === 'measurement' ? 1100 : 200), `${id}: ${words} default narrative words`);
  }
 });
@@ -28,17 +28,18 @@ test('mobile story diagrams preserve the same nodes and directed relationships',
    const name = /<ArchifyDiagram name="([^"]+)"[^>]+ story/.exec(page)[1];
    const desktop = JSON.parse(read(`diagrams/${name}.architecture.json`));
    const mobile = JSON.parse(read(`diagrams/${name}-mobile.architecture.json`));
-    assert(desktop.components.length >= 5 && desktop.components.length <= 7, name);
+    assert(desktop.components.length >= 5 && desktop.components.length <= (name === 'strategy-target' ? 18 : 7), name);
-    const semantic = d => ({nodes:d.components.map(({pos,size,...n})=>n),edges:d.connections.map(({fromSide,toSide,...e})=>e)});
+    const semantic = d => ({nodes:d.components.map(({pos,size,...n})=>n),edges:d.connections.map(({fromSide,toSide,via,labelAt,...e})=>e)});
    assert.deepEqual(semantic(desktop), semantic(mobile), name);
-    assert(desktop.connections.every(e => e.label && e.from !== e.to), name);
+    assert(desktop.connections.every(e => (e.label || e.id.startsWith('govern-')) && e.from !== e.to), name);
    assert.equal(desktop.meta.legend.mode, 'hidden');
  }
 });
-test('all sixteen exact KPI formulas remain reachable in the closed reference', () => {
+test('all sixteen exact KPI formulas remain reachable in the linked reference', () => {
-  const page = read('docs/strategy/measurement.mdx');
+  assert(read('docs/strategy/measurement.mdx').includes('./reference/measurement-contracts.mdx'));
-  const contracts = page.split('<summary>All 16 KPI contracts')[1].split('</details>')[0];
+  const page = read('docs/strategy/reference/measurement-contracts.mdx');
  const contracts = page.split('## Metric contracts')[1].split('## Run the synthetic')[0];
  assert.equal((contracts.match(/^### /gm) || []).length, 16);
  for (const metric of model.kpis) {
    const section = contracts.split(`### ${metric.name}\n`)[1]?.split('\n### ')[0];
@@ -3,9 +3,9 @@ import assert from 'node:assert/strict';
 import {readFileSync, readdirSync} from 'node:fs';
 const read = p => readFileSync(p, 'utf8');
 const model = JSON.parse(read('research/ai-adoption/strategy-model.json'));
-test('eight fixed strategy pages preserve target boundary and separate evidence', () => {
+test('seven canonical strategy pages preserve target boundary and separate evidence', () => {
-  assert.equal(model.pages.length, 8);
+  assert.equal(model.pages.length, 7);
-  assert.equal(readdirSync('docs/strategy').length, 8);
+  assert.equal(readdirSync('docs/strategy').filter(p => p.endsWith('.mdx')).length, 7);
  for (const {id} of model.pages) {
    const page = read(`docs/strategy/${id}.mdx`);
    assert(page.includes(`id: ${id}`));
@@ -19,7 +19,7 @@ test('eight fixed strategy pages preserve target boundary and separate evidence'
  }
 });
 test('every metric has an actionable reader-facing contract', () => {
-  const page = read('docs/strategy/measurement.mdx');
+  const page = read('docs/strategy/reference/measurement-contracts.mdx');
  assert.equal(model.kpis.length,16);
  for (const k of model.kpis) {
    const section=page.split(`### ${k.name}\n`)[1]?.split('\n### ')[0];
@@ -18,6 +18,9 @@ assert(workflow.includes("gitea.event_name == 'push' && gitea.ref == 'refs/heads
 if (process.argv.includes('--build')) {
  const model = JSON.parse(read('research/ai-adoption/strategy-model.json'));
  const indexes = readdirSync('build').filter(f => /^search-index.*\.json$/.test(f)).map(f => read(join('build',f))).join('\n');
  const redirect = read('build/strategy/index.html');
  assert(redirect.includes('/eldorado-v1/documentation/strategy/target-operating-model/'));
  assert(redirect.includes('window.location.href') && redirect.includes('http-equiv="refresh"') && !redirect.includes('archify-panel'), 'Landing must redirect, not duplicate Target');
  for (const {id, title} of model.pages) {
    const route = `strategy/${id === 'index' ? '' : `${id}/`}index.html`;
    const page = read(join('build',route));
@@ -1,2 +1,2 @@
 import type {SidebarsConfig} from '@docusaurus/plugin-content-docs';
-export default {strategySidebar: ['index', 'target-operating-model', 'skills-governance', 'skills-quality', 'skills-adoption', 'measurement', 'skills-maintenance', 'roadmap']} satisfies SidebarsConfig;
+export default {strategySidebar: ['target-operating-model', 'roadmap', 'skills-governance', 'skills-quality', 'skills-adoption', 'measurement', 'skills-maintenance']} satisfies SidebarsConfig;
@@ -110,7 +110,7 @@ a:focus-visible, button:focus-visible {outline: 3px solid #b88719; outline-offse
 .archify-panel--story {max-width: 740px;}
 .archify-panel--story > iframe {height: auto; aspect-ratio: 480 / 408;}
 .archify-panel--story > .story-mobile {display: none;}
-.archify-panel--story[data-diagram="strategy-target"] > iframe {aspect-ratio: 480 / 628;}
+.archify-panel--story[data-diagram="strategy-target"] > iframe {aspect-ratio: 1620 / 1430;}
 .archify-panel--story[data-diagram="strategy-adoption"] > iframe {aspect-ratio: 545 / 408;}
 .archify-panel--story[data-diagram="strategy-maintenance"] > iframe {aspect-ratio: 524 / 408;}
 .strategy-content > details {margin: .65rem 0; border: 1px solid var(--line); padding: .8rem 1rem; background: var(--surface);}
@@ -124,7 +124,7 @@ a:focus-visible, button:focus-visible {outline: 3px solid #b88719; outline-offse
  .archify-panel--story[data-diagram="strategy-roadmap"] > .story-mobile {aspect-ratio: 240 / 628;}
  .archify-panel--story[data-diagram="strategy-adoption"] > .story-mobile {aspect-ratio: 305 / 748;}
  .archify-panel--story[data-diagram="strategy-maintenance"] > .story-mobile {aspect-ratio: 284 / 748;}
-  .archify-panel--story[data-diagram="strategy-target"] > .story-mobile {aspect-ratio: 301 / 868;}
+  .archify-panel--story[data-diagram="strategy-target"] > .story-mobile {aspect-ratio: 680 / 2340;}
 }
 .archify-expand {display: grid; place-items: center; width: 44px; height: 44px; border: 0; padding: 0; background: transparent; color: var(--ifm-link-color); cursor: pointer;}
 .archify-expand:hover {background: var(--ifm-hover-overlay);}
@@ -25,7 +25,7 @@ export default function Home(): React.ReactElement {
        <div className="section-heading"><div><p className="eyebrow">START HERE</p><h2 id="start-title">A clear place to begin.</h2></div><span className="foundation-badge">Foundation · v1</span></div>
        <div className="start-grid">
          <Link className="start-card" to="/organization/structure/"><span className="card-number">01 / ORGANIZATION</span><h3>Organization</h3><p>Explore the structure workspace and its first interactive diagram.</p><span className="card-action">Explore structure <span aria-hidden="true">↗</span></span></Link>
-          <Link className="start-card" to="/strategy/"><span className="card-number">02 / STRATEGY</span><h3>Strategy</h3><p>Shape the direction for an AI-powered Eldorado.</p><span className="card-action">View the direction <span aria-hidden="true">↗</span></span></Link>
+          <Link className="start-card" to="/strategy/target-operating-model/"><span className="card-number">02 / STRATEGY</span><h3>Strategy</h3><p>Shape the direction for an AI-powered Eldorado.</p><span className="card-action">View the direction <span aria-hidden="true">↗</span></span></Link>
          <Link className="start-card" to="/contribution/"><span className="card-number">03 / CONTRIBUTION</span><h3>Contribution</h3><p>Contribute evidence, proposals, documentation, and diagrams.</p><span className="card-action">Start contributing <span aria-hidden="true">↗</span></span></Link>
        </div>
        <aside className="scope-note"><span className="scope-symbol" aria-hidden="true">✦</span><div><strong>Shaping the direction.</strong><p>This space is dedicated to AI strategy and direction for Eldorado. Specific initiatives, priorities, and the roadmap remain to be defined.</p></div></aside>
Author	SHA1	Message	Date
jarvis-at-skic	8dd54f0ab6	docs: explain skills adoption measurement and collection Build and publish Eldorado documentation / build (pull_request) Successful in 39s Details	2026-10-05 18:33:40 +00:00
jarvis-at-skic	8bfcb3f81d	Merge pull request 'fix: canonical Target navigation and Roadmap second' (#21 ) from fix/strategy-navigation into main Build and publish Eldorado documentation / build (push) Successful in 38s Details	2026-10-05 11:19:15 -07:00
jarvis-at-skic	baf79e125b	fix: make Target canonical and place Roadmap second Build and publish Eldorado documentation / build (pull_request) Successful in 40s Details	2026-10-05 18:16:57 +00:00
jarvis-at-skic	9651417ba7	Merge pull request 'docs: label central autonomous agent Human Controlled' (#20 ) from docs/target-human-controlled into main Build and publish Eldorado documentation / build (push) Successful in 40s Details	2026-10-05 11:02:33 -07:00
jarvis-at-skic	4cc2bb4dfd	docs: label central autonomous agent Human Controlled Build and publish Eldorado documentation / build (pull_request) Successful in 1m22s Details	2026-10-05 17:59:59 +00:00
jarvis-at-skic	afb6a97e97	Merge pull request 'docs: center Target lifecycle on autonomous agent' (#19 ) from docs/target-central-agent into main Build and publish Eldorado documentation / build (push) Successful in 40s Details	2026-10-05 10:25:47 -07:00
jarvis-at-skic	1057433c2a	docs: center Target lifecycle on autonomous agent Build and publish Eldorado documentation / build (pull_request) Successful in 41s Details	2026-10-05 17:23:56 +00:00
jarvis-at-skic	504d9f8845	Merge pull request 'docs: lead Strategy with autonomous SDLC cycle' (#18 ) from docs/autonomous-sdlc-cycle into main Build and publish Eldorado documentation / build (push) Successful in 42s Details	2026-10-05 09:34:42 -07:00
jarvis-at-skic	29beebb379	docs: lead Strategy with autonomous SDLC cycle Build and publish Eldorado documentation / build (pull_request) Successful in 51s Details	2026-10-05 16:32:51 +00:00
jarvis-at-skic	e95b85afee	Merge pull request 'fix: right-align inline diagram controls' (#17 ) from fix/right-align-inline-controls into main Build and publish Eldorado documentation / build (push) Successful in 45s Details	2026-10-05 05:51:58 -07:00
jarvis-at-skic	d7e1ad1c0c	fix: right-align inline diagram controls Build and publish Eldorado documentation / build (pull_request) Successful in 48s Details	2026-10-05 12:50:12 +00:00
jarvis-at-skic	714090a83d	Merge pull request 'fix: restore inline diagram controls and synchronize portal theme' (#16 ) from fix/inline-diagram-controls into main Build and publish Eldorado documentation / build (push) Successful in 42s Details	2026-10-05 05:22:41 -07:00
`@@ -1,2 +1,2 @@`
	`import type {SidebarsConfig} from '@docusaurus/plugin-content-docs';`	`import type {SidebarsConfig} from '@docusaurus/plugin-content-docs';`
	`export default {strategySidebar: ['index', 'target-operating-model', 'skills-governance', 'skills-quality', 'skills-adoption', 'measurement', 'skills-maintenance', 'roadmap']} satisfies SidebarsConfig;`	`export default {strategySidebar: ['target-operating-model', 'roadmap', 'skills-governance', 'skills-quality', 'skills-adoption', 'measurement', 'skills-maintenance']} satisfies SidebarsConfig;`