Help us improve
Share bugs, ideas, or general feedback.
From plugin-lab
Review an MCP plugin against the homelab canonical plugin spec and identify any and all misalignments. Use when the user wants a plugin audited for spec drift, wants to compare a plugin to canonical manifests and runtime patterns, wants to know whether deviations are documented and justified, or wants a detailed plugin review report.
npx claudepluginhub jmagar/claude-homelab --plugin plugin-labHow this skill is triggered — by the user, by Claude, or both
Slash command
/plugin-lab:review-lab-pluginThe summary Claude sees in its skill listing — used to decide when to auto-load this skill
Review a plugin for alignment with the homelab canonical plugin spec.
Audits Claude Code plugins for security vulnerabilities, best practices, CLAUDE.md compliance, marketplace readiness, git hygiene, performance, and UX. Produces scored reports using scans for secrets, dangerous commands, and structure validation.
Audits Claude Code plugins for manifest validity, component organization, namespace compliance, documentation, and marketplace readiness before releases or periodic checks.
Audits Claude Code plugins for structure validation, frontmatter quality, deprecations, feature adoption, security patterns, and documentation. Ensures changelog compatibility and best practices for releases.
Share bugs, ideas, or general feedback.
Review a plugin for alignment with the homelab canonical plugin spec.
Find mismatches between the target plugin and the current canonical standard.
Treat the review as a spec audit, not a style pass.
Focus on:
See references/canonical-spec.md for the authoritative correct values for every check category below.
Inspect the target plugin's tree first to know what exists and what is absent. Then read each canonical file when present:
README.mdpyproject.toml, Cargo.toml, or package.json).claude-plugin/plugin.json.codex-plugin/plugin.json.mcp.json.app.jsonCLAUDE.mdAGENTS.mdCHANGELOG.mdDockerfiledocker-compose.yamlJustfile.env.example.github/workflows/ci.yaml)Also inspect the repo tree so you can reason about missing or misplaced surfaces.
Check for every category below. For the exact correct values in each category, consult references/canonical-spec.md.
/health endpoint*_help tool or action/subaction patternFor each finding, state:
Do not treat every difference as a bug. A deviation is acceptable if it is deliberate, documented, and technically justified.
Some findings cannot be resolved by reading files alone. Flag these explicitly as open questions — issues that require user input or live service access to resolve before remediation can proceed.
Examples of open questions:
Group open questions at the end of the report so the user can answer them before align-lab-plugin begins work.
Write the final review to:
docs/reports/plugin-reviews/<YYYYMMDD-HHMMSS>.md
Use the fill-in-the-blanks template in references/review-report-template.md for the report structure.
The report must include:
Findings first. Each finding uses this format:
**[HIGH] Missing /health endpoint**
File: src/server.py
Expected: GET /health returns HTTP 200 with {"status": "ok"}
Found: No /health route registered
Fix: Add health route before starting server
Severity levels:
Keep the report detailed and complete enough that align-lab-plugin can implement all fixes without repeating the audit.