agentleFS
Sign inStart free
sam-mountainmansam-mountainman/codex-mcp/skills/codex-consult/SKILL.md

codex-consult

Skill0 starsChanged 23 days agoMITView on GitHub

Consult OpenAI GPT/Codex from Claude Code for planning, deep reasoning, continuation, or code review. Use when the user explicitly asks GPT, Codex, GPT-5.6, Sol, Terra, or Luna to plan, answer, or review; or asks to continue a prior codex-mcp session.

1 other project uses a version of this.

Adopted by

Most recently taken up by

  • Other versions1 project uses a version of this.

    Word for word · 0

    None found.

    Edited · 0

    None found.

    Wrote their own version · 1

    1. h-wata/claude-skills★ 0same name

    How matches are found. Word for word: the exact same text. Edited: shares passages with this file (“99% same” means 99% of its five-word phrases also appear here; 10% or more counts). Their own version: similar meaning, or, weakest, the same name. We can’t tell which project had it first. Projects are counted among the 200 most-starred candidates, and mentions are links to the repository we found, up to 50 per site, so both counts are a floor.

  • In this repositoryOne other file it gives its agents.

The file

We split this skill into the rules it gives an agent

5 rules. About 0 are common, 1 is distinctive, and 4 are very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈1

    Distinctive · estimate

    This author's own choices. Worth a look.

  • 4

    Very specific

    Tied to their project's names, files or tools. Probably skip.

  • Common
  • Distinctive
  • Very specific
  1. Very specific:

    Use the codex MCP tools supplied by this plugin.

    line 8
  2. Very specific:

    Follow-up in the same external conversation: call gpt_continue with the exact returned session_id.

    line 16
  3. Distinctive:

    Do not call Codex for every ordinary request.

    line 18
  4. Very specific:

    When neither setting is specified, omit both.

    line 40
  5. Very specific:

    All Codex calls run with approval_policy=never and sandbox_mode=read-only.

    line 48

≈ “Common” is an estimate: at least 1 in 5 of the 350,495 skill files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.

Keep what fits and drop the rest. Your version, ready for your agents.
Files
SKILL.md

Codex Consult

Use the codex MCP tools supplied by this plugin. Codex is the read-only external architect/reviewer; Claude Code remains the implementation agent.

Route

  • Setup or authentication issue: call gpt_status first.
  • Implementation design: call gpt_plan, then implement the useful parts with Claude Code.
  • Difficult question or second opinion: call gpt_ask.
  • Review after implementation: call gpt_review.
  • Follow-up in the same external conversation: call gpt_continue with the exact returned session_id.

Do not call Codex for every ordinary request. Use it when the user explicitly asks for GPT/Codex, requests this plugin, or asks for an external quality pass.

Model and reasoning selection

Pass a model only when the user names one. Map common names as follows:

  • GPT-5.6 -> gpt-5.6
  • GPT-5.6 Sol or Sol -> gpt-5.6-sol
  • GPT-5.6 Terra or Terra -> gpt-5.6-terra
  • GPT-5.6 Luna or Luna -> gpt-5.6-luna

Pass reasoning_effort only when the user specifies a level or intent:

  • quick, light, fast, 軽く, サクッと -> low
  • balanced, normal, 普通 -> medium
  • deep, careful, 深く, じっくり -> high
  • very deep, 徹底的 -> xhigh
  • maximum, max, 本気で, 最大 -> max
  • ultra, delegation, 並列で徹底的 -> ultra

Explicit values win over these mappings. Valid values exposed by the MCP server are none, low, medium, high, xhigh, max, and ultra. The local Codex model catalog is authoritative; if a combination fails, call gpt_status for that model and report the supported levels.

When neither setting is specified, omit both. The server default is gpt-5.6-sol with max, configurable through CODEX_MCP_DEFAULT_MODEL and CODEX_MCP_DEFAULT_REASONING_EFFORT.

Relay and continuation

Preserve the external model's concrete findings, caveats, and file references. You may synthesize them with your own judgment, but clearly distinguish disagreements.

For a follow-up, reuse the exact session_id. Changing the model or reasoning level on gpt_continue is allowed when the user requests it.

All Codex calls run with approval_policy=never and sandbox_mode=read-only; never imply that Codex changed the repository.

Download now (ZIP)

Does it work?

Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.

No reviews yet

People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.

  • It worked
  • Partly
  • It didn't
What did you use it for? e.g. Reviewing pull requests in a Rails app

Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.

Your trail

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.

Now do this for your own.

Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.

Check your setup, free