/autoreview

name: autoreview
description: "Structured Codex, Claude, Amp, Pi, or Kimi code review when explicitly requested."

Auto Review

Run an independent review when the user or an owning workflow asks for one. This is code review, not Guardian approval routing. Let the reviewer choose how to analyze the change; provide the target, relevant context, and desired severity. Findings are advice to verify, not instructions to apply blindly.

Run

Use `scripts/autoreview` beside this skill. Keep its custom `codex exec` path: native `codex review` cannot combine explicit Git target flags with custom instructions. The helper combines those with evidence, severity filtering, and validated JSON; it leaves review judgment to Codex. For an OpenClaw checkout:

AUTOREVIEW=".agents/skills/autoreview/scripts/autoreview"
"$AUTOREVIEW" --mode local

In the canonical agent-skills repo, the path is `skills/autoreview/scripts/autoreview`. On Windows, invoke the helper with Python. Use `--help` for the complete flags and environment overrides.

Choose the Git target explicitly when the default is ambiguous:

| Target | Arguments | Scope | | ------------------------------ | ------------------------------ | ----------------------------------------------------------- | | Local work | `--mode local` | HEAD → index → working tree, plus untracked files | | Local candidate against a base | `--mode local --base <ref>` | Pinned base → index → working tree, plus untracked files | | Committed branch/PR | `--mode branch --base <ref>` | Merge-base → HEAD; excludes dirty work | | One commit | `--mode commit --commit <ref>` | Raw parent → commit; a root compares against the empty tree |

`--mode auto` selects local work when dirty, otherwise a branch review using the PR base or `origin/main`. Clean main has no implicit review target. `--mode uncommitted` is an alias for local. The helper does not fetch refs.

Registered nested linked checkouts from the same repository are outside the current review scope. Their presence or edits do not make the parent dirty; ordinary adjacent files remain included in the review. Worktree boundaries are revalidated without changing Git ignore rules.

For a complete PR candidate **including dirty rewrites**, use local mode with its pinned merge base—not branch mode:

pr_base=$(gh pr view --json baseRefName --jq .baseRefName)
merge_base=$(git merge-base HEAD "origin/$pr_base")
"$AUTOREVIEW" --mode local --base "$merge_base"

When a file has both staged and unstaged changes, both states are reviewed. A defect in the index remains actionable even if the working tree fixes it; the report labels it `INDEX-only`. Git display settings cannot suppress context markers or add patch colors; repository configuration is not changed. Source paths and text retain literal whitespace. An empty present source uses line 1, column 1, and an empty excerpt; empty physical lines also use an empty excerpt at column 1. Source identity remains mandatory.

Local selection honors `core.autocrlf` from external operator Git configuration, with repository-local values and attributes retaining precedence. Only its validated scalar value reaches diff/status; other global and system Git configuration stays disabled. Repository-owned or relative global-config overrides are not imported, and reviewed source bytes are not rewritten.

Context and severity

Use `--prompt` for task-specific guidance, or `--prompt-file` and `--dataset` for repository-relative context files. Context does not expand the selected Git target. The reviewer cannot read unchanged repository files from its empty sandbox; supply relevant source or dependency evidence when the diff is insufficient.

The default threshold is **P0 only**: material blockers to normal operation or safety. Use `--max-priority P1`, `P2`, or `P3` when the caller requests a wider review. Do not add unrelated redesign goals or prescribe file counts, reading sequences, or ritual extra passes. Historical blame requires a verified parent-relative patch; otherwise leave the attribution unknown.

"$AUTOREVIEW" --mode local --prompt-file review-notes.md --dataset evidence.json

Engines

Codex is the default: `gpt-5.6-sol`, high reasoning, with a `gpt-5.6-terra` retry only for an account-access failure. Honor explicit engine/model choices; do not switch because a review is slow or rate-limited.

Use `--engine`, `--model`, and `--thinking` to override the defaults. `--codex-speed fast` selects priority service when supported. Only Claude accepts `--fallback-model`. Per-engine environment overrides use `AUTOREVIEW_<ENGINE>_*`.

By default, Codex preserves only authentication settings from user configuration; provider, profile, context and catalogue settings remain ignored. To project a named route, select it explicitly through the existing config override:

"$AUTOREVIEW" --mode local --codex-config 'model_provider="review_api"'

The selector must match `model_provider` in the operator's external `CODEX_HOME/config.toml`. It accepts one bare or simply quoted identifier; provider definitions and other capabilities cannot be supplied through overrides. Projection requires Python 3.11 or `tomli`; default auth-only operation retains its existing fallback parser.

The selected route must use `https://api.openai.com/v1` and command authentication with an absolute external executable. Fixed arguments belong in that executable's wrapper; omitted or empty `auth.args` are accepted. Omitted `wire_api` and `requires_openai_auth` retain Codex's `responses` and `false` defaults. Optional auth timing and context settings keep native defaults and semantics.

On POSIX, a private launcher restores the validated caller `HOME` only for the selected authe