# OpenAI Codex CLI

> OpenAI Codex CLI is an officially published AI tool from OpenAI. Anomity scores its risk at 10 out of 100 (low).

- **Canonical URL:** https://catalog.anomity.ai/agent/codex-cli
- **Kind:** AI tool
- **Publisher:** OpenAI
- **Trust level:** official
- **Risk score:** 10/100 (low)
- **Website:** https://github.com/openai/codex
- **Last updated:** 2026-09-02
- **JSON:** https://catalog.anomity.ai/api/entries/agent/codex-cli

## Description

OpenAI terminal coding agent with sandboxed execution

## Why it scored 10

The score starts at the catalog's neutral base of 50 and moves only through the signals below. The formula is published at https://catalog.anomity.ai/about — there is no model and no hidden heuristic.

| Signal | Contribution | Severity | Evidence |
| --- | --- | --- | --- |
| Published by the vendor | -40 | trust offset |  |

Clamped result: **10/100**, band **low**.

## Configuration surface

Files this tool reads on the host. Anything written here changes what the agent does next.

- `codex-global-config` — Global configuration (TOML)
    - win32: `${USERPROFILE}\.codex\config.toml`
    - darwin: `${HOME}/.codex/config.toml`
    - linux: `${HOME}/.codex/config.toml`
- `codex-managed-system-config` — System config layer (below user config)
    - darwin: `/etc/codex/config.toml`
    - linux: `/etc/codex/config.toml`
- `codex-managed-prefs` — MDM machine policy (macOS managed preferences, base64 TOML payloads)
- `codex-managed-requirements` — Admin-enforced requirements (allowed approval policies / sandbox modes)
    - win32: `${PROGRAMDATA}\OpenAI\Codex\requirements.toml`
    - darwin: `/etc/codex/requirements.toml`
    - linux: `/etc/codex/requirements.toml`
- `codex-managed-defaults` — Managed default settings (user-writable on Windows — claimed, not enforced)
    - win32: `${USERPROFILE}\.codex\managed_config.toml`
    - darwin: `/etc/codex/managed_config.toml`
    - linux: `/etc/codex/managed_config.toml`
- `codex-user-profiles` — Named profile configs layered over the base user config
- `codex-global-instructions` — Global agent instructions
    - win32: `${USERPROFILE}\.codex\AGENTS.md`
    - darwin: `${HOME}/.codex/AGENTS.md`
    - linux: `${HOME}/.codex/AGENTS.md`
- `codex-global-instructions-override` — Global agent instructions override (preferred over AGENTS.md)
    - win32: `${USERPROFILE}\.codex\AGENTS.override.md`
    - darwin: `${HOME}/.codex/AGENTS.override.md`
    - linux: `${HOME}/.codex/AGENTS.override.md`
- `codex-project-config` — Project-level configuration
    - win32: `${PROJECT}\.codex\config.toml`
    - darwin: `${PROJECT}/.codex/config.toml`
    - linux: `${PROJECT}/.codex/config.toml`
- `codex-project-instructions` — Project-level agent instructions
    - win32: `${PROJECT}\AGENTS.md`
    - darwin: `${PROJECT}/AGENTS.md`
    - linux: `${PROJECT}/AGENTS.md`
- `codex-project-instructions-override` — Project-level agent instructions override (preferred over AGENTS.md)
    - win32: `${PROJECT}\AGENTS.override.md`
    - darwin: `${PROJECT}/AGENTS.override.md`
    - linux: `${PROJECT}/AGENTS.override.md`
- `codex-global-hooks` — Global lifecycle hooks (PreToolUse / PostToolUse / SessionStart / etc.)
    - win32: `${USERPROFILE}\.codex\hooks.json`
    - darwin: `${HOME}/.codex/hooks.json`
    - linux: `${HOME}/.codex/hooks.json`
- `codex-project-hooks` — Project-level lifecycle hooks
    - win32: `${PROJECT}\.codex\hooks.json`
    - darwin: `${PROJECT}/.codex/hooks.json`
    - linux: `${PROJECT}/.codex/hooks.json`
- `codex-auth` — Plaintext credential storage (when cli_auth_credentials_store="file")
    - win32: `${USERPROFILE}\.codex\auth.json`
    - darwin: `${HOME}/.codex/auth.json`
    - linux: `${HOME}/.codex/auth.json`
- `codex-plugin-manifest` — Codex plugin manifests (per cached plugin)
- `codex-plugin-mcp` — MCP servers bundled with installed Codex plugins
- `codex-plugin-hooks` — Lifecycle hooks bundled with installed Codex plugins
- `codex-plugin-skill` — Skills bundled inside installed Codex plugins
- `codex-user-skill` — Personal Codex skills (~/.agents/skills/)
- `codex-home-skill` — Personal Codex skills in the Codex home itself (~/.codex/skills/) — the sibling of ~/.agents/skills that newer Codex builds read
- `codex-project-skill` — Project-scope Codex skills (.agents/skills/)
- `codex-user-subagent` — User-defined subagents (~/.codex/agents/*.toml)
- `codex-project-subagent` — Project-scoped subagents (.codex/agents/*.toml)
- `codex-user-rules` — User-level Starlark rules (~/.codex/rules/*.rules)
- `codex-project-rules` — Project-level Starlark rules (.codex/rules/*.rules)
- `codex-model-catalog` — Custom model catalogs (~/.codex/model-catalogs/*.json)

## Related AI tools

- [AdaL](https://catalog.anomity.ai/agent/adal) — risk 10 (low)
- [Agent Zero](https://catalog.anomity.ai/agent/agent-zero) — risk 10 (low)
- [Aide](https://catalog.anomity.ai/agent/aide) — risk 10 (low)
- [Aider](https://catalog.anomity.ai/agent/aider) — risk 10 (low)
- [AiderDesk](https://catalog.anomity.ai/agent/aider-desk) — risk 10 (low)
- [Amazon Q Developer](https://catalog.anomity.ai/agent/amazon-q) — risk 10 (low)

---

Source: Anomity Catalog (https://catalog.anomity.ai/). Scoring methodology: https://catalog.anomity.ai/about. Machine-readable index: https://catalog.anomity.ai/llms.txt · https://catalog.anomity.ai/openapi.json