codex

Files (1)

• SKILL.md

SKILL.md (first lines)

# Codex

Codex is an independent agent on PATH (`codex` — invoke as `command codex` if
a shell alias shadows it), sharing this working tree and already authenticated. It is a second opinion, not ground truth: verify what it
reports, own what it changes. It reads the same skills your repo carries.

Leave Codex's model at its CLI default unless the user asks for a specific
one. The default follows the current Codex release; do not pin a dated model
ID in this skill. If an explicit override is needed, verify it is current and
pass the requested model with `-m <model>`.

## If `codex` is not installed

When `codex` is missing from PATH, offer to install it — **ask the user for
approval first**, never install on your own initiative. On yes, follow the
current instructions at https://developers.openai.com/codex/cli. First-run
authentication is interactive — hand that step to the user. Verify with
`codex --version` before proceeding.

## Prompting Codex

Prompt Codex like an operator, not a collaborator: compact, block-structured
with XML tags. State the task, what "done" looks like, and the few constraints
that matter. A tighter prompt beats a bigger run — improve the contract before
raising `--effort`.

- **One task per run.** Split unrelated asks (review, then fix, then docs) into
  separate runs; a mixed prompt gets a mixed result.
- **Name skills instead of restating them.** Codex reads the same skills your
  repo carries — say "follow write-docs for the doc", "obey refactor-clean: no
  compatibility wrappers." Don't re-explain what a skill already carries.
- **Blocks, added only where the task needs them:**
  - `<task>` — the concrete job, the repo/failure context, the expected end
    state. Nearly always present.
  - `<output_contract>` — exact shape, highest-value first, compact.
  - `<default_follow_through>` — take the low-risk interpretation and keep
    going; stop only when a missing detail changes correctness, safety, or an
    irreversible action.
  - `<verification_loop>` — before finalizing, check the result against the
    requirements and the changed files; revise rather than ship the first
    draft. Any risky fix.
  - `<grounding>` — ground every claim in code or tool output; label inferences
    as inferences. Review and research.
  - `<action_safety>` — keep the diff tightly scoped; no drive-by refactors.
    Write tasks.
- **Anti-patterns:** vague framing ("take a look"), no output contract ("report
  back"), "think harder" in place of a contract, mixing jobs in one run, and
  demanding certainty the evidence can't support.

## Review — proactive

Run a Codex review whenever you have a substantive diff you'd want a second set
of eyes on — a refactor, a tricky algorithm, renderer work, a security-sensitive
change — before declaring it done or committing. Skip it for trivial edits
(typos, comments, doc-only).

1. Fix the diff scope before launching: `--uncommitted` for working-tree
   changes, `--base <branch>` for a branch diff, `--commit <sha>` for a landed
   commit. Resolve a landed commit to its full SHA. Use the non-interactive
   review entry point with an explicit read-only sandbox and separate progress
   and verdict files in a fresh temporary directory:

   ```sh
   review_dir=$(mktemp -d "${TMPDIR:-/tmp}/codex-review.XXXXXX")
   codex exec --sandbox read-only -C "$(git rev-parse --show-toplevel)" review \
     --commit "$(git rev-parse HEAD)" --json -o "$review_dir/verdict.md" \
     < /dev/null > "$review_dir/events.jsonl" 2> "$review_dir/stderr.log"
   ```

   Background long runs and retain their process/session handle. Scope flags
   and custom instructions are mutually exclusive: a scope flag takes no
   prompt. For custom framing, specify the exact diff to inspect in the prompt;
   don't rely on default scope or traverse all refs/checkpoint history. Never
   state the answer you expect (unprimed, same discipline as
   [screenshot-critique](../../visual/screenshot-critique/SKILL.md)).
2. Monitor progress in `events.jsonl` and failures in `stderr.log`. An active
   reviewer reading relevant code is not hung merely because five minutes
   elapsed. If it stalls or drifts outside the diff and its dependencies,
   inspect the last command/error before interrupting; correct the cause or
   narrow the task before retrying. Completion requires exit status zero,
   `turn.completed`, and a nonempty final verdict. A timeout, `turn.failed`, or
   missing verdict is incomplete, never a clean review.
3. Triage every finding: confirm it against the code before acting. Preserve
   Codex's evidence boundaries — an inference it labelled is not a fact.
   Overlap with your own doubts is high-priority evidence; a finding you
   dismiss needs a stated reason, not silence.
4. Report the outcome to the user — what Codex flagged, what you fixed, what
   you dismissed and why. Done when every finding is either fixed or
   explicitly dismissed.

## Implementation — explicit ask only

Delegate implementation to Codex only when the user names Codex for the task.
Never hand it work on your own initiative, and never re-delegate follow-up
work without a fresh ask.

1. Slice the task sharp before delegating — goal, constraints, and how to
   verify — using the prompt discipline above. An underspecified task stays
   with you until a fresh agent couldn't misread it.
2. Start from a clean tree (or record the baseline commit) so Codex's diff is
   separable from yours.
3. Pick the sandbox by what the task must RUN:
   - Pure code + typecheck/unit: `codex exec --sandbox workspace-write "<task>"`.
     Network is off; add `-c sandbox_workspace_write.network_access=true` only
     when the task must fetch (e.g. new deps).
   - **Browser verification, dev servers, or full test runs: use
     `codex exec --dangerously-bypass-approvals-and-sandbox "<task>"`.** The
     sandbox blocks localhost binds (`listen EPERM` on vite/playwright), so a
     sandboxed codex ships code it never saw run. Bypass trades that blindness
     for zero OS control: only in a dedicated git worktree, only with a prompt
     you authored end-to-end (never relaying third-party text), and the diff
     review you owe afterwards is the control.
   Use `-o <file>` to capture the final message and background long calls;
   suppress codex's stderr thinking-noise with `2>/dev/null` so it doesn't
   bloat your context (drop it only to debug a failing run), and add
   `--skip-git-repo-check` to run outside a git repo. Non-interactive runs
   never ask for approval either way.
4. Follow up with `codex exec resume <session-id> "<follow-up>"`, taking the
   id from the run header. `resume --last` means the most recent session

From dzhng/skills, MIT.

More from dzhng/skills