Deepen Skill Body Parity behind one seam

[derek-palmer]

This was generated by AI from an Architecture Review (arch-review).

What to build

One rule — "a distributed skill copy's body (frontmatter stripped) must equal the canonical skill body" — is currently implemented three times: the installer owns strip_frontmatter/extract_frontmatter and does the parity compare in its install planning; scripts/validate_skill_copies.py re-implements the same strip + compare; and the doctor health check dynamically loads that validation script by file path (via importlib + a uuid-suffixed module name) just to borrow strip_frontmatter. The installer's own docstring admits the coupling ("Body extraction matching scripts/validate_skill_copies.py"). Change one, the other two silently drift.

Promote a single Skill Body Parity module that owns frontmatter stripping and the canonical-vs-copies comparison, sourcing artifact paths from the Distribution Inventory. The installer's install planning, the doctor parity check, and the skill-copy validator all consult it instead of re-deriving the logic. The doctor's importlib/uuid script-loading seam is deleted.

This is a deepening refactor: observable behavior (install body-parity abort, doctor finding, validator exit code) is unchanged — only the owner of the rule moves.

Acceptance criteria

• A single module owns frontmatter stripping and canonical↔copies body comparison, consuming Distribution Inventory paths.
• Installer install-planning, doctor parity check, and scripts/validate_skill_copies.py all call this module; no duplicate strip/compare implementations remain.
• Doctor no longer dynamically loads scripts/validate_skill_copies.py (no importlib/uuid script loader) for body parity.
• Body parity is covered by direct unit tests (canonical + copies in, result out) — not only through the doctor --run-scripts path or a full installer write.
• Existing installer, doctor, and validator behavior/tests pass unchanged; full suite green.

Blocked by

None - can start immediately.

[coderabbitai]

🔗 Related PRs

#8 - docs: add agent plugin distribution roadmap [merged]
#11 - feat(agent): add canonical skill source [merged]
#13 - Feat/claude skill package [merged]
#14 - docs: sync all markdown to v0.3.2 reality [merged]

---

📝 Issue Planner

_{Check the box below or use the @coderabbitai plan command to generate an implementation plan and prompts that you can use with your favorite coding assistant.}

• Create Plan

---

🧪 Issue enrichment is currently in open beta.

You can configure auto-planning by selecting labels in the issue_enrichment configuration.

To disable automatic issue enrichment, add the following to your .coderabbit.yaml:

issue_enrichment:
  auto_enrich:
    enabled: false

💬 Have feedback or questions? Drop into our discord!