菜单
Iterative code review loop against a GitHub PR — atk-code-review posts findings as self-review threads, this skill fixes, replies, and resolves them, up to 10 passes.
用 Codex 或 Claude 帮你安装 复制这段 Prompt，粘贴到 Codex、Claude 或其他助手里，让它检查 Skill 页面并帮你完成安装。
基于 SOC 职业分类
Write a "megaplan" planning notebook — a long-form Markdown doc for a multi-slice project that tracks scope, decisions, requirements, risks, and grounded research facts with cross-referenced IDs. Use when the user asks to write up a project plan, planning notebook, or "megaplan" doc, especially one spanning multiple tickets/PRs where decisions and raw research need to be preserved for later reference. Also use when bootstrapping a new megaplan from a vague idea. Use this skill every time you work on a `*.megaplan.md` file, or a Markdown file that has a work plan with tickets.
Strengthen a plan by running it through a subagent review loop — catching inaccuracies and oversights before execution.
Improve a plan by asking a subagent to generate an alternative perspective, then comparing and combining results. Use when a plan already exists in the conversation and the user wants a second opinion, broader coverage, or to catch blind spots.
Help the user develop a vague idea into a "plan" — a scoped, handoff-ready statement of intent
When the user has PR review feedback to address, CI failures to fix, or wants to automatically resolve common PR issues. Use when the user says "fix the PR," "address review comments," "CI is failing," "fix the build," "autofix," "pr feedback," "resolve comments," "fix merge conflicts," "babysit the PR," or when a PR needs updates before merge. Reads the review comments, diagnoses issues, and applies fixes.
Review code changes and return a prioritised list of issues — posted as GitHub PR review threads when a PR exists, or as plain text otherwise. Use when asked to review a PR, diff, or implementation for correctness, code quality, reuse, efficiency, safety, or test quality. When invoking, pass available details - the plan, git change range, files to review.
| name | polish-implementation |
| description | Iterative code review loop against a GitHub PR — atk-code-review posts findings as self-review threads, this skill fixes, replies, and resolves them, up to 10 passes. |
Run an iterative review loop on an implementation, fixing and replying to self-review threads on GitHub each pass.
This skill pushes commits and posts/resolves GitHub PR comments autonomously across passes — same posture as atk-pr-autofix. Invoking it is the opt-in; it won't ask before each push or comment.
Ensure the PR is up to date.
git push (first run on a new branch: git push -u origin HEAD).gh pr view --json number -q .number 2>/dev/null.gh pr create --draft.atk-code-review reviews it directly, no separate diff-scoping step needed here.Subagent review. Spawn a subagent against the PR from step 1 and the plan (if available), using the atk-code-review skill. It gathers PR context, dedups against existing threads (any author), and posts a single batched self-review to the PR — returning the new thread IDs it created (empty if nothing new was found).
If no new threads: done — go to step 6 (report).
Triage and fix. For each thread ID returned by step 2:
atk-pr-autofix's categories: must fix / should fix / won't fix / needs discussion.addPullRequestReviewThreadReply (takes threadId + body directly, no comment-id lookup needed): quote the finding, state the action taken and commit hash (or the reason for not fixing).bash <SKILL_DIR>/scripts/resolve-review-threads.sh [number] --thread-id <id>[,<id>...]atk-pr-autofix's job to handle later.Push, then return to step 2. Repeat up to 10 passes.
Report.
Notes:
Reporting:
$polish-implementation run — there may be more to find.