Hook Claude Code
Runs when a session starts on startup, resume, clear and compact, executing toolkit-local-bridge.cjs via node with --hook, --sync-enabled and --write. From weijunswj/ai-agent-toolkit.
Hook Claude Code
Runs before the agent uses a tool for Agent and Task tool calls, executing toolkit-claude-agent-hook.cjs via node. From weijunswj/ai-agent-toolkit.
Plugin Claude Code
Plugin marketplace listing 1 plugin: ai-agent-toolkit.
Plugin Claude Code
Source-first AI-agent toolkit with native Claude Code plugin metadata, English setup routing, Windows-safe hook repair, opt-in repo auto-update, and local bridge adapters.
Instructions file CodexOpenCode
AGENTS.md instructions for weijunswj/ai-agent-toolkit, covering ai agent toolkit repo rules, ai coding agent rules, instruction priority, working modes and agent topology and delegation.
Instructions file
Claude Code instructions for weijunswj/ai-agent-toolkit, a project described as: AI Agent Toolkit plugin for AI-assisted coding governance, tools and skills.
Instructions file Gemini CLI
Gemini CLI instructions for weijunswj/ai-agent-toolkit, a project described as: AI Agent Toolkit plugin for AI-assisted coding governance, tools and skills.
Skill Claude CodeCodex
Use when working with n8n workflows in any capacity. The always-on protocol for the n8n-skills plugin, loaded by the SessionStart hook every session. Routes to the right skill, summarizes every n8n MCP tool (closing the deferred-description gap), and lists the cross-cutting rules.
Skill Claude CodeCodex
Audit third-party AI agent skills, SKILL.md folders, skill packs, or GitHub skill repositories before converting them into a toolkit. Use when asked to review, import, port, strip unsafe behavior from, preserve provenance for, or decide whether to convert Claude/Codex/Copilot/OpenClaw/Antigravity/Cursor/Gemini-style…
Skill Claude CodeCodex
Use when asked to install, check, repair, refresh, or bootstrap repo-local agent rules. Bootstrap or repair repo-local AI coding agent instruction files such as AGENTS.md, CLAUDE.md, GEMINI.md, or .agents/rules. Use once for unchecked target folders, then stop unless instruction-file state may have changed.
Skill Claude CodeCodex
Use when creating or maintaining source-traceable AI-facing repo surfaces, portable skills, MCP notes, templates, pack metadata, manifests, source locks, audit baselines, or anti-drift documentation.
Skill Claude CodeCodex
Explicit-only current-main N5 GitHub parent/direct-child governance and truthful PR-review/Deferred Findings reconciliation.
Skill Claude CodeCodex
Use when reviewing or planning local AI stack setup such as local LLM runners, model servers, Stable Diffusion or ComfyUI-style web UIs, vector databases, model downloads, GPU/runtime changes, or exposed local AI endpoints. Do not use for ordinary AI API coding, model prompting, or general app work.
Skill Claude CodeCodex
Use when reviewing an implementation plan to compare low-cost managed or owner-hosted foundations before custom-building auth, accounts, databases, backend APIs, workflow automation, CRM/contact pipelines, forms, email, storage, analytics, ops, traffic/security monitoring, or account-security foundations.
Skill Claude CodeCodex
Use when working on any n8n task, including official n8n Skills, using-n8n-skills, n8n workflow JSON, official n8n MCP, n8nlive, workflow creation, workflow updates, helper scripts, import/export, validation, credentials, webhook IDs, workflow activation, execution, repo/live sync, and n8n safety. Always apply before…
Skill Claude CodeCodex
Use when setting up safe n8n environments: localhost/ngrok dev stack, production Cloudflare Tunnel self-hosting for local/CGNAT machines, hosted n8n on Hostinger Coolify VPS, production backup templates, launcher/menu use, official n8n Skills setup, or instance-level MCP references. For workflow, helper-script, or…
Skill Claude CodeCodex
Use when selecting, reviewing, or copying public reusable n8n workflow JSON templates that must be generic, inactive, credential-free, and safe for publication, after applying n8n-agent-rules.
Skill Claude CodeCodex
Use when running a guarded final audit, completion audit, production-readiness audit, release-candidate audit, launch-readiness audit, QA pass, "make sure everything works", "is this production ready", "/goal" readiness remediation, security-readiness check, or final readiness check. Performs preflight only until…
Skill Claude CodeCodex
Use when planning security-first CI/CD setup, GitHub Actions workflows, pipelines, pipeline hardening, required checks, security checks, deployment automation, build/test automation, automated validation, secret scanning, branch protection, release automation, installer prompt/status templates, or safe rollout…
Skill Claude CodeCodex
Use when reviewing or planning non-n8n self-hosted service setup, Docker Compose services, VPS deployment notes, reverse proxies, public ports, DNS/TLS, tunnels, admin panels, backups, or first-run hardening. Do not use for ordinary app coding, local-only dev servers, or n8n work.
Skill Claude CodeCodex
Use when the user says "setup toolkit" or "refresh toolkit", asks to repair the installed n8n Skills plugin or fix n8n .sh hooks opening in an editor on Windows, or when the task is clearly about AI Agent Toolkit plugin setup/update state, Toolkit Local Bridge setup or troubleshooting, repo-backed auto-update, bridge…
Skill Claude CodeCodex
Use when creating, reviewing, or improving frontend web interfaces, including design systems, landing pages, SaaS dashboards, forms, component planning, accessibility, responsive polish, privacy-safe UX, and implementation review.
Skill Claude CodeCodex
Use when starting, relaunching, verifying, or debugging Windows localhost web apps, API servers, or dev services, especially long-running dev servers, bounded readiness polling, interrupted command cleanup, PowerShell startup, execution policy, Corepack/package-manager, Python runtime, spawn EPERM, duplicate…