Agent skill
expert-panel
ACTIVATE when debugging a skill, analyzing why a prompt isn't working, or stuck on a technical problem. Triggers on "what would the panel say", "let's debug this", "why isn't this working", "convene the panel", or when user expresses confusion about skill behavior.
Install this agent skill to your Project
npx add-skill https://github.com/credentum/vivarium-lab/tree/main/persona-skills
SKILL.md
Expert Panel Analysis
CRITICAL: Activation Signal
When this skill activates, you MUST begin your response with exactly:
[PANEL CONVENED] experts take their seats
Process
Analyze the current problem through five expert lenses:
1. Prompt Engineer
Focus: Signal clarity, system prompt competition Ask: Is the description unambiguous? Is this fighting other instructions?
2. Skills Builder
Focus: Discovery, loading, structure Ask: Did it load? Right directory? Restart needed? Structure correct?
3. Character/Roleplay Person
Focus: Performability, anchoring Ask: Can this be performed? Are there hooks to grab? Verbal tics?
4. Interpretability Researcher
Focus: Mechanism, baseline behavior Ask: Fighting trained behavior? How strong is the steering needed?
5. Pragmatist
Focus: Observability, simplicity Ask: Can we measure it? What's the simplest test? What are we actually debugging?
Output Format
For each expert:
- State their likely diagnosis (1-2 sentences)
- Key question they'd ask
- Whether this lens reveals the issue (Yes/No/Partial)
Then: Recommended Action based on panel consensus.
Example
Problem: "My custom skill isn't triggering"
[PANEL CONVENED] experts take their seats
Prompt Engineer: "Description may be too vague or competing with other triggers."
- Key Q: What exact phrases should activate this?
- Reveals issue: Partial
Skills Builder: "Might not have loaded at all."
- Key Q: Can Claude list this skill when asked directly?
- Reveals issue: Yes ✓
Character Person: "Can't assess until we know it's loading."
- Key Q: N/A
- Reveals issue: No
Interpretability: "N/A until skill is in context."
- Key Q: N/A
- Reveals issue: No
Pragmatist: "Start with the simplest test - is it even there?"
- Key Q: Run
ls ~/.claude/skills/and ask Claude to list skills - Reveals issue: Yes ✓
Recommended Action: Verify skill discovery before debugging content. Check directory structure and restart Claude Code.
Recommended Agent Skills
Expand your agent's capabilities with these related and highly-rated skills.
mattpocock/skills
git-guardrails-claude-code
Set up Claude Code hooks to block dangerous git commands (push, reset --hard, clean, branch -D, etc.) before they execute. Use when user wants to prevent destructive git operations, add git safety hooks, or block git push/reset in Claude Code.
mattpocock/skills
setup-pre-commit
Set up Husky pre-commit hooks with lint-staged (Prettier), type checking, and tests in the current repo. Use when user wants to add pre-commit hooks, set up Husky, configure lint-staged, or add commit-time formatting/typechecking/testing.
mattpocock/skills
migrate-to-shoehorn
Migrate test files from `as` type assertions to @total-typescript/shoehorn. Use when user mentions shoehorn, wants to replace `as` in tests, or needs partial test data.
mattpocock/skills
handoff
Compact the current conversation into a handoff document for another agent to pick up.
mattpocock/skills
obsidian-vault
Search, create, and manage notes in the Obsidian vault with wikilinks and index notes. Use when user wants to find, create, or organize notes in Obsidian.
mattpocock/skills
edit-article
Edit and improve articles by restructuring sections, improving clarity, and tightening prose. Use when user wants to edit, revise, or improve an article draft.
Didn't find tool you were looking for?