SKILL.md

@@ -208,7 +208,7 @@ Architecture owns complete Task decomposition for every Feature allocated to the
 2. define at least two independently executable Tasks per Feature unless a recorded Architecture rationale proves that one atomic Task is the smallest verifiable boundary;
 3. map the Feature's Tasks collectively to every acceptance outcome and every applicable requirement, with reciprocal canonical traceability;
 4. encode within-Feature sequencing and every cross-Feature prerequisite as acyclic Task dependencies;
-5. give each Task one specific outcome, bounded scope, repository/component owner, future acceptance-evidence requirement, and canonical/display identity;
+5. give each Task one specific outcome, repository/component owner, canonical/display identity, task-specific implementation artifacts, concrete inputs and outputs, failure boundaries, exclusions, verification steps, and future acceptance-evidence requirements;
 6. keep draft Tasks separate from implementation approval and Kanban Focus admission.
 
 Architecture must not report planning readiness while any Feature allocated to the release lacks this complete Task breakdown. A Feature-shaped placeholder Task, a generated empty Task group, or requirement coverage without executable Tasks does not satisfy the gate.

tests/test_task_breakdown_policy.py

@@ -24,7 +24,12 @@ class TaskBreakdownPolicyTest(unittest.TestCase):
         required = [
             "version: 1.7.0",
             "repository/component owner",
-            "future acceptance-evidence requirement",
+            "acceptance-evidence requirements",
+            "implementation artifacts",
+            "inputs and outputs",
+            "failure boundaries",
+            "exclusions",
+            "verification steps",
             "within-Feature sequencing",
             "cross-Feature prerequisite",
             "acyclic Task dependencies",