merge-ready

GitHub

将代码分支推进至可合并状态,执行同步、多轴子代理审查(安全、计费、复杂度等)、修复问题并提交PR,确保通过检查且保留最终人工审核权。

.agents/skills/merge-ready/SKILL.md VladSez/easy-invoice-pdf

Trigger Scenarios

用户要求将功能/修复/分支变为 merge-ready 请求构建并使更改准备好用于审查

Install

npx skills add VladSez/easy-invoice-pdf --skill merge-ready -g -y
More Options

Non-standard path

npx skills add https://github.com/VladSez/easy-invoice-pdf/tree/main/.agents/skills/merge-ready -g -y

Use without installing

npx skills use VladSez/easy-invoice-pdf@merge-ready

指定 Agent (Claude Code)

npx skills add VladSez/easy-invoice-pdf --skill merge-ready -a claude-code -g -y

安装 repo 全部 skill

npx skills add VladSez/easy-invoice-pdf --all -g -y

预览 repo 内 skill

npx skills add VladSez/easy-invoice-pdf --list

SKILL.md

Frontmatter
{
    "name": "merge-ready",
    "description": "Take a branch from \"code exists (or is about to)\" to \"ready for Ben's final review\" — multi-axis subagent review with verified findings, fixes, ci:check, checkpoint commits, and an updated PR. Use whenever the user says a feature\/fix\/branch should be \"merge ready\", asks to get changes ready for review, or appends this to a build request (\"build X and make it merge-ready\")."
}

Merge ready

Drive the current work to the point where the only remaining step is Ben's own review and merge. The deliverable is a pushed branch with a clean pnpm ci:check, checkpoint commits along the way, and an open PR with a high-level description plus review instructions.

Never merge the PR. Ben always reviews last.

0. Figure out the starting point

This skill composes with feature work — it is not only a review pass:

  • Invoked alongside a build request ("build X, make it merge-ready"): implement the feature/fix first, committing as you go, then continue below. The review phases cover all changes on the branch vs origin/main, not just the last edit.
  • Invoked on existing work ("make this branch merge-ready"): start directly at step 1. The scope is git diff origin/main...HEAD plus anything uncommitted.

1. Sync with main

  • git fetch origin main. If the branch is behind, merge origin/main in and resolve conflicts (favor main's version for code this branch didn't intentionally change).
  • Checkpoint: commit the merge before starting review, so conflict resolution is auditable separately from review fixes.

2. Multi-axis subagent review

Spawn independent review subagents in parallel, one per axis, each given repo access and the complete branch scope:

  • committed changes: git diff origin/main...HEAD
  • staged changes: git diff --cached
  • unstaged changes: git diff
  • untracked files: git status --short, followed by reading every in-scope untracked file

Do not let an uncommitted or newly created file escape review merely because it is absent from origin/main...HEAD.

  1. Unnecessary complexity — thin wrappers, needless indirection, single-use abstractions, defensive guards for impossible states, dead config. This codebase deliberately stays simple.
  2. Security — authz on new endpoints (org/project scoping), SSRF, injection, secrets handling, anything user-input-shaped reaching D1/R2/external APIs.
  3. Billing & metering — ways a user could trigger DataForSEO/provider spend without being metered, charged-but-failed paths, retry/loop amplification, endpoints with unexpectedly high per-call user cost. Credits are billed via Autumn; uncounted spend is a revenue leak.
  4. Library & project idioms — TanStack (Router/Query/Start) used idiomatically; patterns match how the rest of the codebase already does it (shared application/provider error boundaries, db/schema conventions, existing component patterns). Flag novel patterns where an established one exists.
  5. Vibe-coded cruft — leftover scaffolding, stale comments narrating the edit history, console.logs, TODO-without-owner, copy-pasted near-duplicates, files/exports nothing uses.

Each reviewer returns findings with file:line, severity (blocker / should-fix / nitpick), and a one-line rationale. Tell reviewers explicitly: this is an early-stage product — do not chase theoretical edge cases; mark anything debatable as nitpick.

3. Verify findings — never blindly accept

For each blocker and should-fix finding, spawn verification subagents (in parallel) that adversarially check the finding against the actual code and verdict APPLY / APPLY-MODIFIED / REJECT with reasoning. Drop rejected findings. Nitpicks don't need verification — they're reported, not necessarily fixed.

Preserve review learnings

After verification, route durable learnings without forcing every review to change policy:

  • If an APPLY or APPLY-MODIFIED finding reveals a recurring or high-risk repository invariant that existing .greptile/ context and CI do not capture, use maintain-greptile-rules and apply its promotion bar.
  • Keep one-off bugs as code fixes and regression tests. Put deterministic mechanical checks in CI or lint instead of Greptile.
  • When a small tooling, documentation, or workflow frustration occurs, use papercuts to append it to .agents/PAPERCUTS.md; do not derail merge-ready work to fix it.

4. Fix, check, loop

  • Apply verified blocker/should-fix fixes. Apply nitpicks only when trivial and clearly right; otherwise list them in the PR for Ben to judge.
  • Checkpoint: commit fixes in logical groups (e.g. one commit per axis or per concern) so the fix history is reviewable on its own.
  • Run pnpm ci:check (prettier, knip, tsc, oxlint). Fix failures and re-run until clean. If a fix was substantial (not formatting/lint), run a quick re-review of just that change.
  • Loop until ci:check passes and no verified findings remain unaddressed.

5. Push and open/update the PR

  • Push the branch. Open a PR against main if one doesn't exist; otherwise update the existing PR's description.
  • PR description requirements:
    • High-level — what changed and why, written for a human skimming. No file paths, no per-file changelog.
    • How to review — a short ordered guide: what to look at first, what the risky/judgment-call areas are, what was deliberately left out of scope.
    • Review notes — unfixed nitpicks and any REJECT verdicts worth a second opinion, clearly labeled as such.
  • Report back to Ben: PR link, one-paragraph summary, and anything that still needs his judgment. Do not merge.

Version History

  • c393236 Current 2026-08-20 10:57

Same Skill Collection

.agents/skills/caveman/SKILL.md
.agents/skills/changelog-maintenance/SKILL.md
.agents/skills/competitive-landscape/SKILL.md
.agents/skills/competitor-analysis/SKILL.md
.agents/skills/emil-design-eng/SKILL.md
.agents/skills/frontend-design/SKILL.md
.agents/skills/keyword-clustering/SKILL.md
.agents/skills/keyword-research/SKILL.md
.agents/skills/link-prospecting/SKILL.md
.agents/skills/make-interfaces-feel-better/SKILL.md
.agents/skills/openseo-release-notes/SKILL.md
.agents/skills/seo-coach/SKILL.md
.agents/skills/seo-project-setup/SKILL.md
.agents/skills/vercel-composition-patterns/SKILL.md
.agents/skills/vercel-react-best-practices/SKILL.md
.agents/skills/web-design-guidelines/SKILL.md
.agents/skills/writing-guidelines/SKILL.md
.cursor/skills/trigger-cost-savings/SKILL.md
.agents/skills/programmatic-seo/SKILL.md
.agents/skills/seo-audit/SKILL.md
.agents/skills/vercel-react-view-transitions/SKILL.md
.cursor/skills/trigger-authoring-chat-agent/SKILL.md
.cursor/skills/trigger-authoring-tasks/SKILL.md
.cursor/skills/trigger-getting-started/SKILL.md

Metadata

Files
0
Version
c393236
Hash
a7124fa4
Indexed
2026-08-20 10:57

inicio - Wiki
Copyright © 2011-2026 iteam. Current version is 2.155.2. UTC+08:00, 2026-08-22 07:30
浙ICP备14020137号-1 $mapa de visitantes$