#!/usr/bin/env python3
from pathlib import Path
import re

ROOT = Path(__file__).resolve().parents[1]
SKILL = ROOT / "SKILL.md"
text = SKILL.read_text(encoding="utf-8")

assert text.startswith("---\n"), "SKILL.md must start with YAML frontmatter"
parts = text.split("\n---\n", 1)
assert len(parts) == 2 and parts[1].strip(), "frontmatter must close and body must be non-empty"
frontmatter, body = parts
assert re.search(r"^name:\s*documentation-docusaurus\s*$", frontmatter, re.M), "wrong skill name"
assert "description:" in frontmatter, "missing description"
assert len(text) <= 100_000, "SKILL.md exceeds Hermes size limit"
assert "## Authoritative repository" in body
assert "## Gitea delivery gate" in body
assert "references/gitea-docusaurus-delivery.md" in text

required = [
    ROOT / "README.md",
    ROOT / "PROVENANCE.md",
    ROOT / "templates/gitea-actions-build.yml",
    ROOT / "references/gitea-docusaurus-delivery.md",
]
assert all(path.is_file() for path in required), "required package artifact missing"

markdown_files = [SKILL, ROOT / "README.md", ROOT / "PROVENANCE.md", *sorted((ROOT / "references").glob("*.md"))]
link_pattern = re.compile(r"\[[^]]*\]\(([^)]+)\)")
missing = []
for source in markdown_files:
    source_text = source.read_text(encoding="utf-8")
    for target in link_pattern.findall(source_text):
        target = target.split("#", 1)[0]
        if not target or "://" in target or target.startswith(("mailto:", "#", "/")):
            continue
        resolved = (source.parent / target).resolve()
        if ROOT not in resolved.parents and resolved != ROOT:
            missing.append(f"{source.relative_to(ROOT)} -> outside package: {target}")
        elif not resolved.exists():
            missing.append(f"{source.relative_to(ROOT)} -> missing: {target}")
assert not missing, "broken local links:\n" + "\n".join(missing)

reference_names = {path.name for path in (ROOT / "references").glob("*.md")}
coverage_text = "\n".join(path.read_text(encoding="utf-8") for path in markdown_files)
unreferenced = sorted(name for name in reference_names if name not in coverage_text.replace(str(ROOT), ""))
assert not unreferenced, f"unreferenced support files: {unreferenced}"

print(
    f"OK skill=documentation-docusaurus chars={len(text)} "
    f"references={len(reference_names)} markdown_files={len(markdown_files)}"
)