One search for your own team's files and the public ecosystem. See how real projects write their agent instructions, and bring the good ideas home.
Files you open will collect here, so you can find them again.
Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.
tmolavillms.txttmolavi/mcp-agent-skills-hub/llms.txt
A curated repository of AI Agent Skills and Model Context Protocol (MCP) server configurations maintained by Taghi Molavi. It contains 277 unique, deduplicated agent skills and setup guides designed for Antigravity, Codex, Cursor, and Claude.
tmolaviReads credentialsSkilltmolavi/agent-project-discovery-skill/SKILL.md
Universal startup and context discovery skill for AI coding agents to inspect, understand, and plan before modifying existing projects.
tmolaviReads credentialsAGENTS.mdtmolavi/agent-project-discovery-skill/templates/AGENTS.md
Welcome, AI Agent. Treat this repository as a production environment maintained by senior engineers. Before proposing changes, executing commands, or asking questions, you MUST follow the Discovery Protocol defined below. "Understand before you change. Inspect before you ask. Decide autonomously in EVERY project." Act like a senior engineer newly onboarded to this codebase. Do not make assumptions, do not rewrite functional patterns, and do not make changes without establishing your mental model first. Universal Scope: All rules in this policy…
tmolaviAGENTS.mdtmolavi/n8n-agent-skills/AGENTS.md
This repository teaches AI coding agents how to build production-grade n8n workflows. Author: Taghi Molavi · Repo: https://github.com/tmolavi/n8n-agent-skills All skills: ./skills/*/SKILL.md Example configs: ./configs/ Tool routing: ./skills/n8n-mcp-tooling/SKILL.md
tmolavillms.txttmolavi/n8n-agent-skills/llms.txt
No summary in the file. Open it to read it.
tmolaviSkilltmolavi/n8n-agent-skills/skills/n8n-agents/SKILL.md
Use when building n8n AI Agent / LangChain clusters — tools, memory, RAG, structured output, system prompts, sub-workflow-as-tool, or fixing agent connection topology (ai_languageModel, ai_tool, ai_memory).
tmolaviSkilltmolavi/n8n-agent-skills/skills/n8n-ai-seo-geo-automation/SKILL.md
Use when building n8n workflows for SEO, AEO, GEO, AI visibility monitoring, content pipelines, schema/JSON-LD checks, rank/citation tracking, or integrating Molavi GEO tools (geo-scope, mcp-geo-server) with n8n.
tmolaviSkilltmolavi/n8n-agent-skills/skills/n8n-binary-data/SKILL.md
Use when handling files, images, PDFs, attachments, binary properties, vision model inputs, or merging binary with JSON context in n8n.
tmolaviSkilltmolavi/n8n-agent-skills/skills/n8n-code-nodes/SKILL.md
Use when considering a Code node, writing JavaScript/Python in n8n, fixing Code return shape errors, or deciding between expression, Edit Fields, and Code.
tmolaviSkilltmolavi/n8n-agent-skills/skills/n8n-credentials-security/SKILL.md
Use whenever auth, API keys, OAuth, tokens, secrets, or credential setup appears in n8n work. Enforces credential system usage, least privilege, and safe HTTP auth patterns.
tmolaviSkilltmolavi/n8n-agent-skills/skills/n8n-data-tables/SKILL.md
Use when designing or querying n8n Data Tables for durable state, deduplication, checkpoints, lightweight CRM-like storage, or idempotency keys inside workflows.
tmolaviSkilltmolavi/n8n-agent-skills/skills/n8n-debugging/SKILL.md
Use when n8n workflows fail, return empty data, drop items silently, show expression errors, or when diagnosing executions with MCP explain/timeline tools.
tmolaviSkilltmolavi/n8n-agent-skills/skills/n8n-error-handling/SKILL.md
Use for production n8n workflows, webhook APIs, retries, error outputs, error workflows, dead-letter patterns, and correct HTTP response status shapes for callers.
tmolaviSkilltmolavi/n8n-agent-skills/skills/n8n-expressions/SKILL.md
Use when writing or debugging n8n expressions (`{{ }}`), `$json`, `$('Node')`, `$input`, Luxon/DateTime, field mapping, or "expression error" / wrong data after Merge.
tmolaviSkilltmolavi/n8n-agent-skills/skills/n8n-loops/SKILL.md
Use for batching, pagination, "for each", Loop Over Items, Split Out, Aggregate, rate limits, or processing arrays in n8n. Default per-item execution often removes the need for an explicit loop node.
tmolaviSkilltmolavi/n8n-agent-skills/skills/n8n-mcp-tooling/SKILL.md
Use when configuring n8n MCP servers, choosing tools across official and community MCPs, routing generate/lint/explain/CRUD calls, or debugging MCP auth. Covers official instance MCP, AutomateLab, czlonkowski, salacoste, leonardsellem.
tmolaviSkilltmolavi/n8n-agent-skills/skills/n8n-node-configuration/SKILL.md
Use when configuring any n8n node parameters — HTTP Request, Webhook, triggers, Merge, Switch/IF, databases, spreadsheets, AI sub-nodes. Never guess typeVersion or property dependencies; load live schema first.
tmolaviSkilltmolavi/n8n-agent-skills/skills/n8n-production-ops/SKILL.md
Use for multi-environment n8n setups, backups, CI lint of workflow JSON, activation policy, secrets rotation, observability, and safe promotion from staging to production.
tmolaviSkilltmolavi/n8n-agent-skills/skills/n8n-subworkflows/SKILL.md
Use when extracting reusable n8n logic, multi-step builds, sub-workflow triggers, tagging for discovery, passthrough vs Define Below inputs, or exposing workflows as agent tools.
tmolaviSkilltmolavi/n8n-agent-skills/skills/n8n-workflow-lifecycle/SKILL.md
Use when starting, designing, organizing, validating, testing, publishing, or handing off any n8n workflow. Covers naming, folders/projects, sticky notes, validation vs verification, side-effect-aware testing, and production handoff.
Plain text files in a repository that tell a coding agent how the project works: commands to run, conventions to follow and things to avoid. CLAUDE.md, AGENTS.md, cursor rules and skills are the common kinds.
CLAUDE.md is read by Claude Code. AGENTS.md is an open format that Codex, Cursor and other agents read. Many projects keep one and point the other at it.
A folder with a SKILL.md that describes one capability, such as filling PDFs or reviewing code. The agent loads it only when the task calls for it.
Your agents already can, over MCP, limited to the files you're allowed to read. Searching them from this page is coming.
