glitchwerks

claude-plugin-authoring

Basics of authoring a Claude Code plugin — the plugin.json manifest, the .claude-plugin/ directory layout, the component types (skills, agents, commands, hooks, MCP), and the validate/install workflow. Use when creating a new Claude Code plugin or wiring one into a marketplace.

0 Updated today

dispatch

Mode-aware dispatch skill for the claude-wayfinder deterministic 7-decision matcher. Real-catalog mode is the default — the skill reads dispatch context JSON from stdin, runs the matcher against your live catalog, and returns the decision JSON verbatim. Pass --demo to opt into bundled fixtures instead. Catalog path resolution: --catalog-path flag > $DISPATCH_CATALOG_PATH env var > canonical default (~/.claude/state/dispatch-catalog.json). If neither --demo nor a resolvable catalog is present the skill emits a [CATALOG ERROR] and exits non-zero.

claude-audit

Audit a project's effective Claude Code configuration — custom and plugin-provided agents and skills — and produce a structured overlap/conflict report with keep / modify / drop recommendations scoped to the project's stated objectives. Trigger this skill whenever the user types `/claude-audit`, asks to "audit my claude config", "find overlap in my agents", "check for skill conflicts", "are any of my agents duplicates", "what's redundant in my setup", or any similar request to review the agent/skill surface for the current project. Also trigger after a fresh plugin install when the user wants to verify nothing new collides with what's already there.

Testing & QA Listed

setup-prospector

Set up claude-prospector's plugin-owned Python venv. Invoked via /setup-prospector or natural-language triggers: "set up claude-prospector", "install prospector dependencies", "prospector isn't working", "fix prospector", "repair prospector". Do not trigger on "dashboard", "usage analysis", or "skill adoption" — those are distinct skills.

refresh-catalog

Manually regenerate the dispatch catalog from skill sidecars and agent frontmatter. Trigger this skill whenever the user types /refresh-catalog, says "regenerate catalog", "refresh dispatch catalog", "rebuild catalog", "update the catalog", or any similar request to force a fresh dispatch catalog build outside of the automatic SessionStart or mtime-check paths.

router-health

Reports the deterministic-dispatch router's health — CI invariants, runtime telemetry, drift events, and catalog state. Wraps `claude-wayfinder health --report` and adds an Analysis section (drill-down on FAIL'ing or near-threshold metrics) plus an extended Notable Findings section (top dispatched agents/skills, catalog freshness). Trigger phrases: "/router-health", "router health", "dispatch health", "router status", "is the router healthy", "router metrics", "show router stats", "router report", "check the router", "router drift".

setup-wayfinder

Materialize the claude-wayfinder Python venv at ${CLAUDE_PLUGIN_DATA}/venv/ and write the setup-state flag. Use when: - User types /setup-wayfinder - User says "set up claude-wayfinder", "install wayfinder dependencies" - User says "wayfinder isn't working", "fix wayfinder", "repair wayfinder" - SessionStart banner indicates setup is required and the user wants to proceed - Plugin version bumped and re-setup needed Do NOT trigger on casual mentions of "wayfinder" without setup/install/fix intent.

dispatch-authoring

Matcher-aware authoring and troubleshooting knowledge for the full dispatch authoring surface — trigger frontmatter, applicable_agents, applicable_skills, routable, and description: phrase-matching. Loaded by any agent (router, code-writer, doc-writer, project-planner, etc.) when the user wants to write, improve, troubleshoot, or understand dispatch configuration. Trigger this skill whenever the user types /dispatch-authoring, asks "how do I write triggers", "how do I make frontmatter for my agent", "what's a good keyword weight", "set up triggers", or says "my agent isn't being dispatched", "this skill never matches", "my frontmatter isn't working", "dispatch isn't picking up", or similar authoring or troubleshooting requests around dispatch configuration. Covers the matcher's seven-decision ladder, scoring math, weight ladder {0.25, 0.5, 1.0}, symmetric Porter2 stemming and the no_stem opt-out, fnmatch path-glob footguns, conflict-pair detection, and the audit-catalog CLI pointer.

Code & Development Listed

usage-analysis

Use when the user asks an interpretive question about their Claude Code token spend — what's interesting, what's anomalous, what to change. This skill produces insights and questions; it does NOT enumerate top consumers (the dashboard does that). Use `usage-dashboard` for "show me the numbers." Use `claude-audit` for "audit my config overlap." Use this skill for "tell me what's interesting / what should I change." Trigger phrases: "am I close to my Claude limit", "how much Sonnet am I using", "where are my Claude tokens going", "what's eating my Claude budget", "why is my Claude spend so high", "what should I change about my Claude setup".

Data & Documents Listed

usage-dashboard

Use when the user explicitly asks to regenerate, rebuild, or refresh the `claude-prospector` token-usage dashboard with no interpretive question attached — runs `python -m claude_prospector dashboard`, writes the HTML file, and reports the output path. Do NOT use when the user is asking what the data means, where their budget stands, or what to change (use `usage-analysis` instead — it regenerates as part of answering). Trigger phrases: "regenerate the usage dashboard", "rebuild my claude-prospector dashboard", "refresh the token dashboard", "regen the prospector dashboard".

session-analysis

Use when the user wants a judgment-level read on whether a Claude Code session stayed on task — did the agent do what was originally asked, and what did it acknowledge skipping. This is the interpretive (LLM) complement to the deterministic `session-audit` CLI: 1a extracts ask-vs-done for free; this skill adds the `Variance` and `What was NOT done` judgment that a parser can't compute, then persists a combined record for drift analysis. Cost: ~1-3k tokens of the current session, paid only when invoked. Opt-in by design — run it selectively, not on every session. Best run in a Sonnet (or stronger) session; the judgment quality depends on it. Use `session-audit` (CLI) for the free deterministic ask/actions extract. Use `usage-analysis` for token-spend insights. Use `claude-audit` for agent/skill config overlap. Use THIS skill for "did this session drift from what I asked". Trigger phrases: "/session-analysis", "did this session stay on task", "analyze session drift", "did the agent do what I asked", "what did t

agent-authoring

Expert guidance for writing, editing, and debugging agent and skill frontmatter, CLAUDE.md harness rules, and SKILL.md files.

Code & Development Listed

refactoring-discipline

Behavior-preserving refactoring discipline. Activates when the task involves restructuring, extracting, or renaming existing code without adding new behavior.