Feature audit
Skill jajupmochi/agent-harness/skills/.task-orchestrator-templates/feature-audit
Linlin's curated AI agent harness configuration: workflow rules, skills, hooks, plugin recommendations, tooling preferences, and project templates. Loadable into any new project so a fresh /init can pick the relevant subset.From the repository description
npx -y skills add jajupmochi/agent-harness --skill feature-auditAssembled from the repository path, not quoted from the project. Check it against their README if it does not work.
One thing to look at
- 1 stars1 stars. Stars are a popularity signal and not a quality one, but at this level it is likely that nobody has read this closely except its author, and you would be relying on your own review.
SKILL.md
2.0 KB, 456 tokens by cl100k_base, as published. Nobody here has run it
feature-audit-template
RESEARCH phase
- Inventory ALL features:
- Skills (`ls skills/*/SKILL.md`)
- Rules (`ls rules/*/RULE.md`)
- Hooks — Claude (`ls hooks/*/README.md`) AND Codex (`cat hooks.json`)
- Scripts (`ls scripts/*.{sh,js}`)
- Templates (`ls templates/*/TEMPLATE_README.md`)
- For each feature, check:
- Does it have a Claude Code implementation? (source)
- Does it have a Codex implementation? (wrapper, or mapping)
- Is the Codex wrapper correct? (tool names mapped, hooks present, auto-load configured)
- Does it work with non-vision models? (visual features especially)
- Is it discoverable? (/skills, /plugins, manifest references)
- Record gaps in audit report
AUDIT MATRIX
For each feature, fill this table:
| Feature | Claude | Codex | Auto-load | Non-vision safe | Status | |---|---|---|---|---| | code-verifier | source skill | wrapper | ✅ YES | ✅ n/a | OK | | research-critic | source skill | wrapper | ✅ YES | ✅ n/a | OK | | verify-visual | chrome MCP | screenshot script | ❌ | ✅ YES | FIXED | | review-gate | Stop hook | hooks.json Stop | n/a | ✅ n/a | FIXED | | ... | ... | ... | ... | ... | ... |
Common gaps found
- Skill has Claude source but no Codex wrapper
- Codex wrapper exists but has no frontmatter (not discoverable)
- Codex wrapper has no allow_implicit_invocation (won't auto-load)
- Codex hook script missing for Claude hook equivalent
- Claude tool name (Write|Edit|Bash) not mapped to Codex equivalent
- Visual feature assumes model vision capability
GENERATE recommendations
For each gap: generate fix recommendation with priority (P0=blocking, P1=important, P2=nice-to-have).
What ships with it
Read from the repository
Just SKILL.md. No reference files, no scripts.