Codex

Codex reads two things, and it wants both: .agents/skills/ for focused workflows it loads on demand, and AGENTS.md for guidance that is always on. Its hook event and reply match Claude Code's, so the guard needs no script of its own - a wiring file points at the same two templates.

Skills

magus agent install .agents/skills

The same install reads your AGENTS.md and prints the managed magus block when it is missing or stale, for you to paste between its markers. Nothing writes that file. Skills explains why, and what magus doctor does with the pasted block afterwards.

Codex discovers skills, AGENTS.md, and MCP servers at task start, so start a new task after changing any of them.

MCP

Register the daemon in your user-level ~/.codex/config.toml, never in the repository:

[mcp_servers.magus]
url = "http://127.0.0.1:7391/mcp"
bearer_token_env_var = "MAGUS_MCP_TOKEN"
enabled = true

magus server start
export MAGUS_MCP_TOKEN="$(magus config token print)"
codex mcp list
magus status --probe=liveness,mcp

codex mcp list confirms configuration; the probe confirms the endpoint is serving. MCP covers dedicated connector tokens, the ChatGPT desktop app, and what to do when mcp.address changes.

Guard hook

Hooks are experimental and off by default. Turn them on in ~/.codex/config.toml:

[features]
codex_hooks = true

Then save this as ~/.codex/hooks.json (or .codex/hooks.json), with the paths pointing at wherever you put your copies of the two templates:

{
  "hooks": {
    "PreToolUse": [
      {
        "matcher": "Bash",
        "hooks": [
          {
            "type": "command",
            "command": "GUARD_AGENT_NAME=codex GUARD_NO_ADVISE=1 sh docs/guides/integrations/agents/magus-guard-command.sh",
            "statusMessage": "magus guard: checking command"
          }
        ]
      },
      {
        "matcher": "Edit|Write",
        "hooks": [
          {
            "type": "command",
            "command": "GUARD_AGENT_NAME=codex GUARD_NO_ADVISE=1 sh docs/guides/integrations/agents/magus-guard-path.sh",
            "statusMessage": "magus guard: checking file"
          }
        ]
      }
    ]
  }
}

GUARD_AGENT_NAME labels the observation magus records; it cannot change a verdict. Everything else about the two scripts is described in Guard hook templates, including the variables that let one implementation serve several hosts.

Notifications

Codex runs a program on its notify setting. Shape the event into the canonical envelope and pipe it to magus session notify, exactly as the other hosts do - see Attention hooks for the envelope and the vocabulary.

Coverage and limits

• Hooks are experimental, off by default, and not available on Windows.
• Reports disagree on whether apply_patch, Edit and Write fire PreToolUse. OpenAI's hooks page says they do; at least one third-party reference says Bash only. Treat the second matcher as provisional and confirm it against the current documentation.
• apply_patch delivers a PATCH in tool_input.command, not a file path, so the declared-output guard wants the Edit/Write matcher rather than apply_patch.
• The command rules are executed against this binary with a real event. The file surface is wired per OpenAI's documentation and has not been executed here.
• Lease capture is UNVERIFIED here. Codex's PreToolUse envelope is the same shape magus already reads, so a lease would be captured by the wiring on Claude Code with the matcher changed - but nothing in Codex's documented tool surface is known to hand a sub-agent a prompt, so there is no matcher to name and none has been executed.

Verify

codex mcp list
magus status --probe=liveness,mcp
magus doctor

doctor's agent skills check grades the installed skills and the pasted AGENTS.md block against the running binary. doctor's guard binary and guard wiring checks report which binary a hook resolves and whether any host config actually invokes a current template.