Skill Claude CodeCodex
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.
Cursor rule Cursor
Guidelines for library versioning.
Instructions file
Claude Code instructions for ArcadeAI/arcade-mcp, covering claude.md, what this is, commands, library dependency graph and versioning rules.
Skill Claude CodeCodex
Add or edit hand-authored prose on an Arcade toolkit reference page by writing curation files under toolkit-docs-generator/curation/. Use when asked to document a toolkit's auth setup, explain an enum or parameter, add a warning to a tool, or fix wording on a toolkit page — anything where the text belongs to a…
Cursor rule Cursor
You are writing documentation for Arcade, an AI platform for developers. Follow these guidelines to maintain a consistent tone and style.
Instructions file CodexOpenCode
AGENTS.md instructions for ArcadeAI/docs, covering agent instructions, quick reference, automated style reviews, style review (on pr open/update) and editorial review (after merge).
Instructions file
Claude Code instructions for ArcadeAI/docs, covering arcade docs, commands, architecture, content authoring and pre-commit hooks.
Plugin Claude Code
AI coding agent workflows: BDD, auto-linting, quality reviews, debugging, and refactoring.
Agent Claude Code
Files Safeword's spooled retro findings to the upstream ArcadeAI/safeword tracker and drains the spool. Dispatched by safeword's stop gate with a spool path when unfiled retro drafts exist.
Agent Claude Code
Performs one read-only, fresh-context degraded review after Safeword exhausts every CLI reviewer route.
Command Claude Code
Report safeword parity drift (pairs + contracts). Safeword-repo only.
Hook Claude Code
Runs when a session starts on compact, executing session-node24.sh, session-bun-check.sh, session-dependency-readiness.ts, session-auto-upgrade.ts, session-safeword-context.ts, session-reply-format.ts, session-version.ts, session-lint-check.ts, session-architecture-heal.ts, session-author-model.ts…
Hook Claude Code
Runs when you submit a prompt, before the agent sees it, executing prompt-timestamp.ts, prompt-questions.ts and prompt-retro-nudge.ts via bun (3 commands). From ArcadeAI/safeword.
Hook Claude Code
Runs when the agent finishes a response, executing stop-quality.ts, stop-reentry.ts, stop-self-report.ts, stop-retro-filing.ts and stop-retro.ts via bun (5 commands). From ArcadeAI/safeword.
Hook Claude Code
Runs before the agent uses a tool for Bash, Edit, Write, MultiEdit, NotebookEdit, Bash, Edit, Write, MultiEdit, NotebookEdit, Bash, Bash, Bash and Bash tool calls, executing pre-tool-dependency-readiness.ts, pre-tool-quality.ts, pre-tool-config-guard.ts, pre-tool-git-bare-fix.sh, pre-tool-architecture-stage.ts and…
Hook Claude Code
Runs after a tool call finishes for Edit, Write, MultiEdit, NotebookEdit, Edit, Write, MultiEdit, NotebookEdit, Bash, Edit, Write, MultiEdit, NotebookEdit, Edit, Write, MultiEdit, NotebookEdit, Edit, Write, MultiEdit, NotebookEdit, Edit, Write, MultiEdit, NotebookEdit and Bash tool calls, executing post-tool-lint.ts…
Hook Claude Code
Runs when a session ends, executing session-cleanup-quality.ts via bun. From ArcadeAI/safeword.
Settings file Claude Code
Agent settings declaring 6 hook events (SessionStart, UserPromptSubmit, Stop, PreToolUse) and 5 allowed tools.
Skill Claude CodeCodex
Run a diff-scoped code audit for architecture, dead code, and test quality. Uses the change from main to focus feature reviews; request a repository audit for whole-codebase discovery.
Skill Claude CodeCodex
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.
Skill Claude CodeCodex
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.
Skill Claude CodeCodex
Inspects SafeWord's production retro collector, durable relay, receipts, and filed GitHub issues. Use when checking whether retros are flowing, investigating a receipt or queue state, or reviewing recent retro submissions. Do not use to submit a retro or mutate production recovery state.
Skill Claude CodeCodex
Kill zombie dev servers and test processes. Use when ports are blocked, processes are hanging, or test runners won't start.
Skill Claude CodeCodex
Close a completed local delivery safely. Use when wrapping up a finished coding session by verifying it, merging only with explicit authority, capturing retrospective learning, and cleaning the exact merged branch and worktree. Do NOT use for cloud-agent tasks, unmerged work, or cleanup without a pull request.