From ai-tooling
Evaluate an AI tool, skill, or plugin against the catalog and ACMM framework before adopting it
How this skill is triggered — by the user, by Claude, or both
Slash command
/ai-tooling:evaluate-toolThe summary Claude sees in its skill listing — used to decide when to auto-load this skill
Evaluate a new AI tool before adding it to your workflow. Prevents tool sprawl by checking for overlap and assessing ACMM-level fit.
Evaluate a new AI tool before adding it to your workflow. Prevents tool sprawl by checking for overlap and assessing ACMM-level fit.
/evaluate-tool <repo-url-or-name>
Fetch the repo description and README:
gh api repos/{owner}/{repo} --jq '.description'
gh api repos/{owner}/{repo}/readme --jq '.content' | base64 -d | head -100
Determine:
Read the catalog at ${CLAUDE_PLUGIN_ROOT}/docs/CATALOG.md and find entries in the same category. For each overlapping tool:
Read ${CLAUDE_PLUGIN_ROOT}/docs/WORKFLOW.md. Determine which ACMM level this tool belongs to:
Check: is the user at or approaching that level? Adopting L5 tools without L3 infrastructure is the "autonomous action without guardrails" anti-pattern.
Output a structured evaluation:
## Tool Evaluation: {name}
**What it does:** {one-liner}
**Problem it solves:** {pain point}
**Type:** {type}
**Category:** {category}
**ACMM Level:** L{n}
### Overlap Analysis
{list overlapping tools from catalog, with comparison}
### Recommendation
- **ADOPT** — fills a gap in the current workflow
- **REPLACE {x}** — better than current tool for this problem
- **SKIP** — overlaps with {x} which is already in the workflow
- **DEFER** — relevant at L{n}, but you're at L{m}. Revisit when {condition}.
### Catalog Entry (if adopting)
{pre-filled table row for CATALOG.md}
If the recommendation is ADOPT or REPLACE, offer to add the entry to ${CLAUDE_PLUGIN_ROOT}/docs/CATALOG.md and update ${CLAUDE_PLUGIN_ROOT}/docs/WORKFLOW.md.
npx claudepluginhub mattbutlerengineering/ai-tooling --plugin ai-toolingGuides creation, editing, and verification of skills for AI coding agents using test-driven development with subagent scenarios. Use when authoring or debugging skills.