Codex Harness MCP

Use this skill when a user wants Codex CLI or another MCP-compatible coding client to work through a harness-engineering loop with explicit execution contracts, persistent local knowledge/RAG, durable traces, structured verification evidence, project-local governance policy, PASS/FLAG/BLOCK governance audits, trace-level observability reports, harness profiles, eval cases/runs, Meta-Harness-lite proposals and promotion decisions, natural-language harness spec export, MCP resources/prompts, compact handoff context, multi-client setup, and completion gates before claiming work is done. Triggers

9 stars 2 forksAI, LLMs & Agents

Install

skills CLI (Claude Code, Cursor, Codex and more)

npx skills add chapzin/codex-harness-mcp --skill codex-harness-mcp

Manual install

git clone --depth 1 https://github.com/chapzin/codex-harness-mcp.git /tmp/codex-harness-mcp
mkdir -p ~/.claude/skills
cp -r /tmp/codex-harness-mcp/. ~/.claude/skills/codex-harness-mcp

Personal skills live in ~/.claude/skills. Use .claude/skills in a project folder to share a skill with your team.

From the SKILL.md

# Codex Harness MCP - Contracts, Memory, Gates

Give Codex and other MCP coding agents a local harness instead of a loose prompt.

`codex-harness-mcp` adds a project-local control plane for Codex CLI and MCP-compatible coding clients: contracts before implementation, persistent knowledge from research and implementation lessons, raw traces, structured verification evidence, project-local governance policy, `PASS/FLAG/BLOCK` governance audits, trace-level observability reports, harness profiles, eval records, Meta-Harness-lite proposals and promotion decisions, natural-language harness spec export, next-step recovery, compact handoff context, and explicit gates before claiming completion.

It is designed for long-running coding, research, audit, refactor, and harness-optimization work where the agent needs to remember what happened, reuse what it learned, and prove the result before saying "done".

Public page: https://skills.sh/chapzin/codex-harness-mcp/codex-harness-mcp

## What this skill gives the agent

- A small contract before implementation.
- A local `.codex-harness/` system of record.
- Persistent project knowledge from research and implementation lessons.
- Trace-backed recovery after failures.
- Verification records without command execution inside the MCP server.
- A governance policy plus `PASS/FLAG/BLOCK` audit for contracts, outputs, raw traces, verification, gates, side effects, and subagent bounds.

Read the full skill on GitHub

Details

Skill name: codex-harness-mcp
Repository: chapzin/codex-harness-mcp
License: MIT
Repo updated: 12 May 2026

FAQ

What does the Codex Harness MCP skill do?

How do I install the Codex Harness MCP skill?

skills CLI (Claude Code, Cursor, Codex and more): npx skills add chapzin/codex-harness-mcp --skill codex-harness-mcp. Skills go in ~/.claude/skills (personal) or .claude/skills (project), and Claude loads them automatically when a task matches the description.

Is Codex Harness MCP popular?

It lives in chapzin/codex-harness-mcp, which has 9 GitHub stars and 2 forks.

Related skills

Humanize

diaiq/claude-skill-humanizer

Humanize AI-generated text to avoid AI detection. Use when the user wants to make ChatGPT, Claude, Gemini, or other AI text sound human and pass GPTZero, Turnitin, ZeroGPT, and other AI detectors.

AI, LLMs & Agents

Cli2skill

2233admin/cli2skill

Turn any CLI or MCP server into an Agent Skill. Use when you want to replace an MCP server with a zero-overhead CLI skill, or generate a skill from any command-line tool's --help output.

AI, LLMs & Agents

Local Falcon

local-falcon/local-visibility-skill

Expert guidance on AI Visibility and Local SEO from Local Falcon, the pioneer of geo-grid rank tracking. Provides deep knowledge on optimizing for AI search platforms (ChatGPT, Gemini, AI Mode, AI Overviews, Grok), local pack rankings, Google Business Profile optimization, and actionable strategies for agencies, enterprises, and SMBs. Includes guidance on using Local Falcon's MCP server for data-driven analysis.

AI, LLMs & Agents

Suno Songwriter

regiellis/suno-songwriter-agent-skill

Write, revise, plan, or adapt songs, lyrics, hooks, and AI music prompts. Use for Suno vocal or instrumental packages, song matrices, multilingual adaptation, genre treatments, reusable artist personas, album plans, Suno settings and iteration, sound samples, and cover-art prompts. Supports original style-inspired writing without copying lyrics. Not for standalone music theory, general DAW mixing, or licensing advice.

AI, LLMs & Agents

Zendesk

andmarios/zendesk-skill

Interact with Zendesk Support via CLI - search tickets, view details, analyze metrics, manage users/organizations, and update tickets. All responses are saved locally for efficient jq querying. All Zendesk content is screened for prompt injection (regex, semantic, and LLM-based) and wrapped with security markers before reaching the LLM.

AI, LLMs & Agents

Self Refine Reflection

Thomaszhou22/self-refine-skill

Systematic self-reflection and iterative output refinement for AI agents. Based on Madaan et al. (2023) Self-Refine, Shinn et al. (2023) Reflexion, and Andrew Ng's Reflection pattern. Use when outputs need multi-round critique-and-improve cycles, when initial quality is insufficient, or when complex tasks benefit from self-correction.

AI, LLMs & Agents

Browse the most popular Claude skills, all ai, llms & agents skills, or write your own with the skills generator.