New: 98,000+ skills indexed from GitHub repos with 100+ stars · every skill safety-checkedBrowse top skillsTop skills
Esc to closeBrowse all skills
Submit a skillBrowse skills
Submit a skillBrowse skills

Agent skill

cmux Pre-Merge Review

by manaflow-ai in manaflow-ai/cmux

Reviews agent-written changes before merge with a review subagent and evidence-backed findings, plus a fuller adversarial protocol for security and data-loss risk.

Custom licenceAuto-check passedDevelopment

Install cmux Pre-Merge Review

skills CLI
$ npx skills add manaflow-ai/cmux --skill cmux-review -a claude-code

Project install by default; add -g for ~/.claude/skills/.

GitHub CLI
$ gh skill install manaflow-ai/cmux cmux-review --agent claude-code

Project scope by default; add --scope user for a personal install. Needs GitHub CLI 2.90.0 or later (public preview).

Manual copy
$ git clone --depth 1 https://github.com/manaflow-ai/cmux.git skills-src && mkdir -p .claude/skills && cp -r skills-src/skills/cmux-review .claude/skills/cmux-review && rm -rf skills-src

Use ~/.claude/skills/ instead of .claude/skills for a personal install. The folder must contain SKILL.md.

Claude Code skills documentation · loads skills from .claude/skills/

Facts

Skill name
cmux-review
GitHub stars
28k
Token cost
~681 tokens
SKILL.md length
312 words
Files
6 (incl. references)
Skills in repo
22
Repo updated
First seen
Licence
Custom licence

At a glance

Reviews agent-written changes before merge with a review subagent and evidence-backed findings, plus a fuller adversarial protocol for security and data-loss risk.

  • Works in 4 steps: Give a review subagent the task intent,… → Verify concrete findings, fix them when… → If fixes were non-trivial, obtain a… → …
  • Reviewing agent-written changes before merging
  • SKILL.md covers Default pre-merge review, Dogfood and merge and Adversarial review
  • Instructions only: no scripts, shell commands, URLs or credentials in SKILL.md

What it does

The default pass gives a review subagent the task intent, the base and head SHAs and the exact diff, and asks for correctness defects first and repository rules second, independent of the author's reasoning. Findings are verified, fixed when authorized, pushed and re-checked with the original failing test, and a green test counts only if it exercises the failure. Non-trivial repairs get a fresh subagent review. Style nits and speculative improvements are suppressed unless requested.

Review stays read-only until an authorized repair, uses subagents in the current runtime rather than a second model or outside service, and does not commit review receipts or scratch files. App, runtime and UI merges need your explicit approval or a direct merge instruction. Security, persistence, concurrency, data-loss or requested deep reviews use an adversarial protocol with source identity capture, independent reviewers, counterevidence, bounded repair and a saved receipt, each covered by a reference file.

When your agent uses it

  • Reviewing agent-written changes before merging
  • Re-reviewing a repair after fixes were pushed
  • Running a deep adversarial review of a security or concurrency change
  • Deciding whether a change is ready to merge

Example prompts

  • “Review the diff on this branch before I merge, and only report real correctness problems.”
  • “Run the adversarial review on the new session persistence code.”
  • “Re-review the fixes I pushed after the last review.”

Requirements

  • An agent runtime that can start review subagents

Workflow steps

4 steps, taken from the first numbered list in SKILL.md.

  1. Give a review subagent the task intent, base/head SHAs and exact diff.
  2. Verify concrete findings, fix them when authorized, and push. Re-run the
  3. If fixes were non-trivial, obtain a fresh subagent review of the repair delta.
  4. Merge when the checks that judge the change pass and the approval rule below

What it can do on your machine

Read from SKILL.md and the folder at commit 1e82fa4. It shows what the files ask for, not the result of running them.

  • Tool permissions

    Pre-approves nothing: there is no allowed-tools line, so your agent's usual permission prompts apply.

    From allowed-tools in the SKILL.md frontmatter.

  • Runs code

    No scripts in the folder and no shell commands in SKILL.md.

    From the folder's file list and the shell code blocks in SKILL.md.

  • Network

    No URLs in SKILL.md.

    From URLs in SKILL.md, links to its own repository left out.

  • Credentials

    Names no API keys, tokens, secrets or passwords.

    From names ending in _API_KEY, _TOKEN, _SECRET, _KEY or _PASSWORD in SKILL.md.

Context cost

cmux Pre-Merge Review loads about 681 tokens when it runs, and up to ~8.1k if it reads all its reference files. Until then it costs about 50 tokens; SKILL.md has 312 words of instructions outside code blocks.

Always · name and description, kept in context so the agent knows when to use it
~50
When it runs · the whole SKILL.md, loaded when a task matches
~681
With references · SKILL.md plus every file in references/, read only if the agent opens them
~8.1k

Estimates: characters ÷ 4, the usual rule of thumb; real counts depend on the model's tokenizer. Scripts and assets cost tokens only if the agent reads them.

Safety

Auto-check passed

The automated check found no risky patterns in SKILL.md.

Automated static check — not a guarantee. Review scripts before installing. It scans the text of SKILL.md for risky patterns (piping downloads into a shell, reading credential files, hidden Unicode, destructive commands); files beside SKILL.md are not scanned.

SKILL.md

Its licence (Custom licence) doesn't allow us to republish the file, so here is its outline and opening line. It has 312 words (~681 tokens).

“Review to reduce developer attention. Surface concrete correctness defects; suppress style, nits and speculative improvements unless requested.”

— opening of SKILL.md by manaflow-ai, Custom licence
name
cmux-review

Read the full SKILL.md on GitHub

Files

SKILL.md and 5 other files (references) in skills/cmux-review of manaflow-ai/cmux.

  • SKILL.md
  • references/dogfood-and-merge.md
  • references/review-discovery.md
  • references/review-receipt.schema.json
  • references/review-receipts.md
  • references/review-verification.md

Open the folder on GitHubat commit 1e82fa4

Compare with similar skills

cmux Pre-Merge Review next to the 5 skills that share the most tags, products or categories with it. Stars are the repository's; “used in” counts other GitHub owners with a copy.

cmux Pre-Merge Review compared with similar skills
SkillStarsUsed inTokensAuto-checkLicenceRepo updated
cmux Pre-Merge Review this skillmanaflow-ai/cmux28k—~681Automated safety check: PassCustom licence
GitHub Review Iterationprisma/orm48k—~2.2kAutomated safety check: PassApache-2.0
Cherry Studio PR ReviewCherryHQ/cherry-studio53k—~3.9kAutomated safety check: PassAGPL-3.0
PR Reviewjaemk/self_update961—~1.5kAutomated safety check: NotesMIT
Agentic Reviewdxos/dxos526—~4.5kAutomated safety check: PassCustom licence
PR Reviewjaemk/cached2.1k—~2.5kAutomated safety check: NotesMIT

Similar skills

  • GitHub Review Iteration

    prisma/orm

    Official

    Runs a loop on a GitHub pull request: fetch review state, triage comments into actions, implement them and resolve threads, repeating until nothing actionable is left.

    48k GitHub stars~2.2k tokensUpdated yesterday
    DevelopmentAuto-check passed

  • Cherry Studio PR Review

    CherryHQ/cherry-studio

    Reviews Cherry Studio branches, pull requests, commits, files and docs against the project's own architecture, naming, API-boundary and UI rules, report-only by default.

    53k GitHub stars~3.9k tokensUpdated today
    DevelopmentAuto-check passed

  • PR Review

    jaemk/self_update

    Targeted, read-only review of a PR or checked-out branch. An agent skill from jaemk/self_update.

    961 GitHub stars~1.5k tokensUpdated 1 mo ago
    DevelopmentAuto-check: notes

  • Agentic Review

    dxos/dxos

    Run the rule-driven agentic code review — discover rule blocks in the repo's .mdl files, prepare per-rule review groups (full project by default; diff-only with --pr-only), spawn one focused Sonnet…

    526 GitHub stars~4.5k tokensUpdated today
    DevelopmentAuto-check passed

  • PR Review

    jaemk/cached

    Targeted, read-only review of a PR or checked-out branch. An agent skill from jaemk/cached.

    2.1k GitHub stars~2.5k tokensUpdated 9 days ago
    DevelopmentAuto-check: notes

More from manaflow-ai/cmux

All 22 skills in this repo

  • cmux Diagnostics

    manaflow-ai/cmux

    Runs a read-only health check for cmux and explains what it finds, covering the CLI and socket, settings, agent hooks, session restore and notifications.

    28k GitHub starsUsed in 1 repo~819 tokens
    Auto-check passed

  • cmux Settings Editor

    manaflow-ai/cmux

    Reads and changes cmux preferences in `~/.config/cmux/cmux.json` through a helper that validates each change before writing, covering terminal, browser, viewers and shortcuts.

    28k GitHub stars~2.3k tokensUpdated today
    Auto-check passed

  • cmux Backend Rules

    manaflow-ai/cmux

    Sets the backend TypeScript and Cloud VM rules for cmux: Effect-based services, thin route handlers, Postgres as source of truth, migrations and provider secrets.

    28k GitHub starsUsed in 1 repo~682 tokens
    Auto-check passed

  • Cmux Debugging Guide

    manaflow-ai/cmux

    Covers debug logging, the Debug menu, profiling rules and runtime pitfalls for working on the cmux macOS terminal app.

    28k GitHub starsUsed in 1 repo~1.1k tokens
    Auto-check passed

  • Ghostty Submodule and GhosttyKit Workflow

    manaflow-ai/cmux

    Workflow rules for the Ghostty submodule in cmux: rebuilding GhosttyKit.xcframework, pushing fork changes and updating the parent submodule pointer safely.

    28k GitHub starsUsed in 1 repo~580 tokens
    Auto-check passed

  • Cmux Release Workflow

    manaflow-ai/cmux

    Runs the cmux release process: gathers per-PR changelog lines since the last tag, bumps the version, guards against a bad tag, then tags and pushes the macOS DMG release.

    28k GitHub starsUsed in 1 repo~856 tokens
    Auto-check passed

Categories

Questions about cmux Pre-Merge Review

What does cmux Pre-Merge Review do?

Reviews agent-written changes before merge with a review subagent and evidence-backed findings, plus a fuller adversarial protocol for security and data-loss risk. The default pass gives a review subagent the task intent, the base and head SHAs and the exact diff, and asks for correctness defects first and repository rules second, independent of the author's reasoning. Findings are verified, fixed when authorized, pushed and re-checked with the original failing test, and a green test counts only if it exercises the failure.

When should I use cmux Pre-Merge Review?

cmux Pre-Merge Review fits situations like: reviewing agent-written changes before merging; re-reviewing a repair after fixes were pushed; running a deep adversarial review of a security or concurrency change; deciding whether a change is ready to merge.

How do I install cmux Pre-Merge Review in Claude Code?

Run `npx skills add manaflow-ai/cmux --skill cmux-review -a claude-code`. Or copy the skill folder (skills/cmux-review in manaflow-ai/cmux) into .claude/skills/cmux-review in your project. Claude Code loads it when a task matches its description.

How do I install cmux Pre-Merge Review in Codex?

Run `npx skills add manaflow-ai/cmux --skill cmux-review -a codex`. Or copy the skill folder (skills/cmux-review in manaflow-ai/cmux) into .agents/skills/cmux-review in your project. Codex loads it when a task matches its description.

Can I use cmux Pre-Merge Review in Cursor, Gemini CLI or GitHub Copilot?

Cursor, Gemini CLI, GitHub Copilot and OpenCode also load SKILL.md folders. With the skills CLI, run `npx skills add manaflow-ai/cmux --skill cmux-review -a cursor` (or -a gemini-cli, github-copilot or opencode for the others). To copy it by hand, put the folder in .cursor/skills/cmux-review, .gemini/skills/cmux-review, .github/skills/cmux-review and .opencode/skills/cmux-review in your project.

What does cmux Pre-Merge Review need to run?

SKILL.md names no scripts, command-line tools or credentials: cmux Pre-Merge Review is instructions for the agent only. Our summary lists: An agent runtime that can start review subagents.

Does cmux Pre-Merge Review access the network?

SKILL.md contains no URLs. Any network use would come from the scripts or tools the agent runs. This is read from the text; nothing was executed.

Is cmux Pre-Merge Review safe to install?

Our automated static check of SKILL.md found no risky patterns, such as piping downloads into a shell, reading credential files or hidden Unicode. It is not a guarantee. Review the folder before installing.

What licence does cmux Pre-Merge Review use?

cmux Pre-Merge Review has a licence file (the repository's licence) that doesn't match a standard licence. Read it on GitHub before reusing the skill.

How many tokens does cmux Pre-Merge Review use?

About 681 tokens (SKILL.md is roughly 2.7k characters). Agents keep only the skill's name and description in context until a task matches; then they load SKILL.md in full. Its references folder adds about 7.4k tokens, read only when the agent opens those files.

What are the alternatives to cmux Pre-Merge Review?

Skills that share tags, products or a category with cmux Pre-Merge Review: GitHub Review Iteration (prisma/orm, 48k stars), Cherry Studio PR Review (CherryHQ/cherry-studio, 53k stars), PR Review (jaemk/self_update, 961 stars) and Agentic Review (dxos/dxos, 526 stars). The comparison table on this page puts their stars, adoption, token cost, safety result and licence side by side.

Who maintains cmux Pre-Merge Review?

manaflow-ai (a GitHub organization) maintains it in manaflow-ai/cmux, which has 28,092 GitHub stars. The repository holds 22 skills in this directory. The repository was last updated on October 11, 2026.

Source: manaflow-ai/cmux on GitHub. Facts on this page come from the repository at the commit we read; the author's words are quoted as theirs.