Spec MVP: project harness audit/design skill suite

[sungjunlee]

Parent: #35

Problem

Maintainers need a practical way to turn a real repo into a better agent workspace without dumping everything into root CLAUDE.md / AGENTS.md.

Proposed MVP

Draft a small skill-suite spec for project harness work. Candidate surfaces:

• craft-harness-audit: inspect an existing repo harness and report what belongs in root context, local context, skills, hooks/scripts, or external tools
• craft-harness-spec: design the target harness shape from the audit plus project constraints

Names are placeholders. The design may choose one skill first if two triggers are premature.

Acceptance Criteria

• The MVP uses progressive disclosure and keeps the core spine short.
• It includes 3-5 eval prompts such as monorepo, legacy repo, small library, multi-agent relay repo, and no-existing-harness repo.
• It defines missing-context behavior: inspect local files first, ask only when authority or project goal cannot be inferred safely.
• It avoids provider-specific implementation in the core spine.

[coderabbitai]

🔗 Related PRs

#3 - Add skill-radar curation layer + craft-skill-spec meta-skill [merged]
#5 - craft-handoff: pair per-session prompt with per-project rich doc [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!