plugins-claude/frontmatter-query/skills/frontmatter-query/SKILL.md
Query YAML frontmatter across markdown files. REQUIRED for discovering and filtering markdown files by metadata — do NOT use raw grep, awk, or sed to parse frontmatter. Use when listing frontmatter fields across files, searching for files by metadata key-value pairs, or counting tag/category distributions.
npx skillsauth add st0nefish/claude-toolkit frontmatter-queryInstall this skill globally with one command. Works with Claude Code, Cursor, and Windsurf.
3 of 9 scanners reported clean
Some scanners were skipped, did not run, or reported a non-clean status. Review each row below.
Use ${CLAUDE_PLUGIN_ROOT}/scripts/frontmatter-query to query YAML frontmatter in markdown files. Do NOT parse frontmatter manually with grep, awk, or sed.
${CLAUDE_PLUGIN_ROOT}/scripts/frontmatter-query list path/to/dir
${CLAUDE_PLUGIN_ROOT}/scripts/frontmatter-query list path/to/file.md
Returns a JSON array of {path, ...metadata} for every .md file with frontmatter.
${CLAUDE_PLUGIN_ROOT}/scripts/frontmatter-query search path/ -k allowed-tools -v Bash
${CLAUDE_PLUGIN_ROOT}/scripts/frontmatter-query search path/ -k tags -v python
Filters to files where the key matches the value. Matching is case-insensitive and supports list membership (e.g., a tags: [python, cli] field matches -v python).
${CLAUDE_PLUGIN_ROOT}/scripts/frontmatter-query tags path/
${CLAUDE_PLUGIN_ROOT}/scripts/frontmatter-query tags path/ -k allowed-tools
Returns {"value": count, ...} sorted by count descending. Defaults to the tags key; use -k to aggregate a different key.
| Flag | Effect |
|------|--------|
| --limit N | Return at most N results |
| --body | Include markdown body in output |
| --keys k1,k2 | Only include these frontmatter keys (plus path) |
frontmatter-query list skills/ to see all skill metadatafrontmatter-query search skills/ -k allowed-tools -v Bash to find skills using BashRead tool on specific files from the resultsCommands using ${CLAUDE_PLUGIN_ROOT}/scripts/frontmatter-query can be auto-approved in Claude Code hooks. This is safe because the script is read-only (stdout output only, no disk writes).
development
Start work from your description — explore the codebase and plan
data-ai
Multi-phase, multi-agent feature workflow: spec → plan → refine → divide → execute → review. Invoke when the user escalates a session-start/session-issue flow to orchestration, or asks to run a non-trivial feature (multiple files, design ambiguity, cross-cutting concerns, correctness-critical paths) through the full multi-agent workflow. For small fixes, prefer session-start.
tools
Browse open issues, pick one, and start work on it
tools
Review, clean up, and open a PR to finalize the work