ArcadeAI

Use when finishing spec.md before writing test-definitions.md, or when the review gate asks for a spec review — self-reviews the just-authored spec inline and earns its Tier 1 review stamp. Your own inline pass; do not spawn a sub-agent.

Safeword semver commitment and release discipline. Use when bumping versions, cutting releases, deciding what goes in a patch vs minor vs major, or reviewing changelog entries. Also use when auto-upgrade logic needs to know what's safe to apply silently.

Run comprehensive code audit for architecture, dead code, and test quality. Use when reviewing overall codebase health, checking for architectural violations, or before marking a feature complete.

Behavior-first feature development — use when building new capabilities, continuing feature work, or when work introduces new state or multiple user flows. Discovers desired behavior through examples and scenarios before implementation. Do NOT use for bug fixes, typos, or small isolated changes.

Use when the user wants to explore options, weigh approaches, or think through uncertainty before committing to a direction. Collaborative brainstorming and rubber ducking — divergence-first thinking partner.

Kill zombie dev servers and test processes. Use when ports are blocked, processes are hanging, or test runners won't start.

Root cause debugging before fixes. Use when investigating bugs, diagnosing test failures, troubleshooting unexpected behavior, or when previous fix attempts failed. Enforces investigate-first discipline.

Extract tacit knowledge through non-obvious microquestions — things only the user knows that can't be found in code, docs, or research. Use when you're about to guess at intent, context, or constraints during SAFEWORD's understanding flow. Also use when user says 'ask me', 'what do you need to know', or when another skill (bdd, brainstorm, debug) needs user context before proceeding. Do NOT use for questions answerable by reading the codebase or searching the web.

Build a custom user verifier server for Arcade.dev in any language. Use when implementing Arcade's user verification flow, building an /auth/verify endpoint, integrating Arcade tool authorization into a production app, or troubleshooting verification issues like missing user_id or session cookies during the OAuth callback.

Build and deploy custom MCP tools using the Arcade MCP framework. Use when the user wants to create, build, scaffold, or deploy an MCP tool or server to Arcade Cloud, or when working with arcade_mcp_server, MCPApp, @tool decorators, arcade deploy, or Arcade tool development.

Build new Arcade error adapters from scratch using public Arcade TDK patterns. Use when adding provider integrations, mapping SDK exceptions, or extending HTTP/GraphQL/auth adapter behavior.

Automate the complete Linear issue workflow - fetch issue, update status to In Progress, create branch, implement, test, create PR, and move to In Review

Deep code quality review with web research. Use when user explicitly requests verification against latest docs ('double check against latest', 'verify versions', 'check security'), needs deeper analysis beyond automatic hook, or is working on projects without SAFEWORD.md/CLAUDE.md. Fetches current documentation (WebFetch), checks latest versions (WebSearch), and provides deep analysis (performance, security, alternatives).