{"slug":"best-skills-for-codex","query":"best skills for Codex","title":"Best skills for OpenAI Codex","url":"https://skills.911fund.io/guides/best-skills-for-codex","updated":"2026-07-13T20:26:48.436Z","kind":"ranked","answer":"Codex reads the open agent-skills standard, so the best Codex tools are the same well-structured, portable skills that lead on other platforms — plus AGENTS.md configurations that give the CLI real project context. Skill of Skills ranks 276 Codex-compatible tools by structural quality rather than stars.","ranking_method":"quality-first composite (structure 55% / reputation 15% / recency 15% / momentum 15)","disclaimer":"Scores are evidence-based quality metadata, not endorsements or install advice. Review each skill's risk level and source before installing.","matched_skills":276,"skills":[{"rank":1,"name":"affaan-m/ECC","slug":"affaan-m-ecc","profile_url":"https://skills.911fund.io/tools/affaan-m-ecc","tier":"curated","quality_score":160,"risk_level":"high","platforms":["claude_code","cursor","codex"],"description":"The agent harness performance optimization system. Skills, instincts, memory, security, and research-first development for Claude Code, Codex, Opencode, Cursor and beyond."},{"rank":2,"name":"graphify","slug":"safishamsi-graphify","profile_url":"https://skills.911fund.io/tools/safishamsi-graphify","tier":"curated","quality_score":165,"risk_level":"high","platforms":["claude_code","codex"],"description":"AI coding assistant skill (Claude Code, Codex, OpenCode, Cursor, Gemini CLI, and more). Turn any folder of code, SQL schemas, R scripts, shell scripts, docs, papers, images, or videos into a queryable knowledge graph. App code + database schema + infrastructure in one graph."},{"rank":3,"name":"openai/openai-agents-python","slug":"openai-openai-agents-python","profile_url":"https://skills.911fund.io/tools/openai-openai-agents-python","tier":"promising","quality_score":109,"risk_level":"medium","platforms":["claude_code","codex"],"description":"A lightweight, powerful framework for multi-agent workflows"},{"rank":4,"name":"cli-printing-press","slug":"mvanhorn-cli-printing-press","profile_url":"https://skills.911fund.io/tools/mvanhorn-cli-printing-press","tier":"curated","quality_score":169,"risk_level":"high","platforms":["claude_code","codex"],"description":"Every API has a secret identity. This finds it, absorbs every feature from every competing tool, then builds the GOAT CLI — designed for AI agents first, with SQLite sync, offline search, and compound insight commands."},{"rank":5,"name":"openai/openai-agents-js","slug":"openai-openai-agents-js","profile_url":"https://skills.911fund.io/tools/openai-openai-agents-js","tier":"promising","quality_score":109,"risk_level":"medium","platforms":["claude_code","codex"],"description":"A lightweight, powerful framework for multi-agent workflows and voice agents"},{"rank":6,"name":"chopratejas/headroom","slug":"chopratejas-headroom","profile_url":"https://skills.911fund.io/tools/chopratejas-headroom","tier":"promising","quality_score":117,"risk_level":"low","platforms":["claude_code","codex"],"description":"Compress tool outputs, logs, files, and RAG chunks before they reach the LLM. 60-95% fewer tokens, same answers. Library, proxy, MCP server."},{"rank":7,"name":"graphify","slug":"graphify-labs-graphify","profile_url":"https://skills.911fund.io/tools/graphify-labs-graphify","tier":"curated","quality_score":175,"risk_level":"high","platforms":["claude_code","codex"],"description":"AI coding assistant skill (Claude Code, Codex, OpenCode, Cursor, Gemini CLI, and more). Turn any folder of code, SQL schemas, R scripts, shell scripts, docs, papers, images, or videos into a queryable knowledge graph. App code + database schema + infrastructure in one graph."},{"rank":8,"name":"openai/plugins","slug":"openai-plugins","profile_url":"https://skills.911fund.io/tools/openai-plugins","tier":"experimental","quality_score":75,"risk_level":"low","platforms":["claude_code","codex"],"description":"OpenAI Plugins"},{"rank":9,"name":"gbrain","slug":"garrytan-gbrain","profile_url":"https://skills.911fund.io/tools/garrytan-gbrain","tier":"curated","quality_score":174,"risk_level":"high","platforms":["claude_code","codex"],"description":"Garry's Opinionated OpenClaw/Hermes Agent Brain"},{"rank":10,"name":"modelcontextprotocol/ext-apps","slug":"modelcontextprotocol-ext-apps","profile_url":"https://skills.911fund.io/tools/modelcontextprotocol-ext-apps","tier":"promising","quality_score":102,"risk_level":"low","platforms":["claude_code","codex"],"description":"Official repo for spec & SDK of MCP Apps protocol - standard for UIs embedded AI chatbots, served by MCP servers"},{"rank":11,"name":"destructive_command_guard","slug":"dicklesworthstone-destructive-command-guard","profile_url":"https://skills.911fund.io/tools/dicklesworthstone-destructive-command-guard","tier":"curated","quality_score":180,"risk_level":"high","platforms":["claude_code","codex"],"description":"The Destructive Command Guard (dcg) is for blocking dangerous git and shell commands from being executed by agents."},{"rank":12,"name":"gastown","slug":"gastownhall-gastown","profile_url":"https://skills.911fund.io/tools/gastownhall-gastown","tier":"curated","quality_score":169,"risk_level":"medium","platforms":["claude_code","cursor","codex"],"description":"Gas Town - multi-agent workspace manager"}],"faqs":[{"q":"What format do Codex skills use?","a":"Codex builds on the open agent-skills (SKILL.md) standard and also reads AGENTS.md for project configuration. Tools authored to these formats are portable across Codex and other compliant clients."},{"q":"Is this list specific to Codex?","a":"It is filtered to Codex-compatible tools, but because the skill standard is cross-tool, many entries also work in Claude Code, Cursor, and other clients."}]}