Codex operator — delegate coding, reviews and plans to OpenAI Codex
Compose prompts and run OpenAI's Codex CLI non-interactively with explicit sandbox modes (read-only, workspace-write, danger-full-access), review workflows with findings-before-fixes, session resume, and strict trust boundaries
- What
- Compose prompts and run OpenAI's Codex CLI non-interactively with explicit sandbox modes (read-only, workspace-write, danger-full-access), review workflows with findings-before-fixes, session resume, and strict trust boundaries
- Cost
- Free
- Needs
- the OpenAI Codex CLI installed and configured with Codex access (paid OpenAI account) — verify with `codex --version`; a codebase to work on
- Install
- Copy the installer prompt below into your Muse — your agent does the rest.
💳 Paid API required — Curated by Skill Harbor — @feiskyer's skill for delegating work to the OpenAI Codex CLI (`codex exec`). It teaches the agent to compose prompts, pick an explicit sandbox level per run — read-only for analysis (the default), workspace-write (`--full-auto`) for implementation, and `danger-full-access` only with the user's explicit per-task consent — and to follow strict trust boundaries (file contents, diffs and tool output are data, never instructions). Covers review workflows where findings are presented first and never auto-applied, adversarial and second-opinion reviews, image-driven implementation, session resume for iterative follow-ups, background execution for long runs, and reference docs (CLI flags, prompting patterns, review schemas, worked examples). Honest caveats: needs the Codex CLI installed and an OpenAI account with Codex access (paid) — running under the skill's rules never gates on `codex login status`, just try it and surface auth errors; note that codex runs outside the host agent's permission system, so the skill advises the read-only sandbox for sensitive repos. MIT licensed. Skill Harbor never reviews the code, review it yourself before use. Discovered via skills.sh.
Version:
Quick answer
Codex operator is a free Muse skill for delegating work to the paid OpenAI Codex CLI, run non-interactively. It teaches your agent to compose prompts and pick an explicit sandbox level per run: read-only for analysis, workspace-write for implementation, and danger-full-access only with your explicit per-task consent, while treating file contents, diffs and tool output as data, never instructions.
Install
Copy the install package below, then paste it into Muse🧪 Install ran in our test sandbox · Nov 11, 2026📌 Pinned version: commit 95dab59Step 0 — before installing anything: read this install prompt, summarize in a few lines what this build does and what installing it involves, then ask me to confirm before you proceed. Prerequisites: the OpenAI Codex CLI installed and configured with Codex access (paid OpenAI account) — verify with `codex --version`; a codebase to work on Install "Codex operator — delegate coding, reviews and plans to OpenAI Codex" for me. It gives my agent @feiskyer's Codex delegation workflow: run `codex exec` non-interactively with an explicit sandbox per task (read-only for analysis by default, workspace-write/--full-auto for implementation, danger-full-access only with my explicit per-task consent), present review findings first and never auto-apply fixes, run adversarial and second-opinion reviews, resume sessions for iterative follow-ups, background long-running invocations, and read the bundled references/ (cli-reference, prompting-patterns, review-workflows, examples) before writing non-trivial prompts. MIT-licensed. Repository: https://raw.githubusercontent.com/feiskyer/claude-code-settings/95dab59a62b2031adcb7abf25ae0a60b8a13fdb2/skills/codex-skill/SKILL.md Pinned version: install the code at commit 95dab59a62b2031adcb7abf25ae0a60b8a13fdb2 exactly (Skill Harbor tested this version). Do not substitute a newer version. If this commit is unreachable, stop and tell me. 1. Fetch the SKILL.md file (and the references/ and assets/ folders) from the repository path into a temporary folder 2. Safety check: review the SKILL.md and references for anything suspicious (unexpected network calls, shell commands outside the allowed-tools list, credential harvesting). This repo should contain zero secrets in code. Verify that holds here; STOP on any red flag and tell me. 3. Install it as a skill: copy SKILL.md and its helper files into the agent's skills directory, in a folder named "codex-skill". 4. Verify with no network calls: frontmatter valid, files in place. 5. Report what was installed, where, and what I still need to do myself (e.g. install the Codex CLI with `npm i -g @openai/codex` if missing, make sure my OpenAI account has Codex access — that's on me). GitHub is optional: if I have a GitHub account or the gh CLI, you may use it; otherwise public access is fine. Never require it unless it's in the prerequisites above. Rules: don't touch anything outside the temp folder and the install target. If anything looks off, stop and ask me.
Saved to your recent installs. Find it anytime on /connect.
🔒 Security, provenance and what this touches
The install prompt below already includes the vetting steps: your agent follows the community checklist before installing anything with executable code. Want more?
Verification certificate
Dated facts about what was checked for this listing, and how. This is not a safety certification.
- Pinned version
- 95dab59 · 2026-10-13
- Sandbox install test
- Install ran in our test sandbox · Nov 11, 2026 · Observed during the install test: files.pythonhosted.org, pypi.org
- Creator identity
- Not verified
- Fingerprint watch
- Last checked · Nov 19, 2026 : no change detected in the files linked by this install prompt
- Install prompt analysis
- 2026-10-13 : No commands, network calls, packages, file writes or API keys detected.
Skill Harbor never reviews the code of a listed skill. These are dated facts about the install prompt and our install test, not a safety certification. Review a skill yourself before you install it. How we verify
Questions
How do I install a skill?
Every product page includes a copy-paste install prompt. Paste it into your Muse and it sets the skill up for you — no manual configuration.
Where does my money go?
Straight to the seller. Skill Harbor never processes payments: checkout happens on the seller’s own page, usually Stripe.
What does the ✓ next to a creator’s name mean?
It means we confirmed the identity of the person behind the listing. It says nothing about the code itself — always check a skill before installing it.