GitHub Review Iteration
prisma/orm
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.
Iterative review loop with subagent reviewer. An agent skill from microsoft/bocpy.
Install Review Loop
$ npx skills add microsoft/bocpy --skill review-loop -a claude-codeProject install by default; add -g for ~/.claude/skills/.
$ gh skill install microsoft/bocpy review-loop --agent claude-codeProject scope by default; add --scope user for a personal install. Needs GitHub CLI 2.90.0 or later (public preview).
$ git clone --depth 1 https://github.com/microsoft/bocpy.git skills-src && mkdir -p .claude/skills && cp -r skills-src/.github/skills/review-loop .claude/skills/review-loop && rm -rf skills-srcUse ~/.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/
Install the "review-loop" agent skill from https://github.com/microsoft/bocpy/tree/main/.github/skills/review-loop into .claude/skills/review-loop/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "review-loop", then confirm the skill loads.Claude Code copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.
$skill-installer install https://github.com/microsoft/bocpy/tree/main/.github/skills/review-loopType this inside Codex. $skill-installer <name> installs a curated skill from openai/skills. The installer writes to $CODEX_HOME/skills (default ~/.codex/skills). Restart Codex if the skill does not show up.
$ npx skills add microsoft/bocpy --skill review-loop -a codexProject install goes to .agents/skills/; add -g for ~/.codex/skills/.
$ gh skill install microsoft/bocpy review-loop --agent codexProject scope by default (.agents/skills/); add --scope user for a personal install.
$ git clone --depth 1 https://github.com/microsoft/bocpy.git skills-src && mkdir -p .agents/skills && cp -r skills-src/.github/skills/review-loop .agents/skills/review-loop && rm -rf skills-srcUse ~/.agents/skills/ instead of .agents/skills for a personal install.
Codex skills documentation · loads skills from .agents/skills/
Install the "review-loop" agent skill from https://github.com/microsoft/bocpy/tree/main/.github/skills/review-loop into .agents/skills/review-loop/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "review-loop", then confirm the skill loads.Codex copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.
$ npx skills add microsoft/bocpy --skill review-loop -a cursorProject install goes to .agents/skills/; add -g for ~/.cursor/skills/.
$ gh skill install microsoft/bocpy review-loop --agent cursorProject scope by default (.agents/skills/); add --scope user for a personal install.
$ git clone --depth 1 https://github.com/microsoft/bocpy.git skills-src && mkdir -p .cursor/skills && cp -r skills-src/.github/skills/review-loop .cursor/skills/review-loop && rm -rf skills-srcUse ~/.cursor/skills/ instead of .cursor/skills for a personal install.
Cursor skills documentation · loads skills from .cursor/skills/, .agents/skills/, .claude/skills/, .codex/skills/
Install the "review-loop" agent skill from https://github.com/microsoft/bocpy/tree/main/.github/skills/review-loop into .cursor/skills/review-loop/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "review-loop", then confirm the skill loads.Cursor copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.
$ gemini skills install https://github.com/microsoft/bocpy.git --path .github/skills/review-loop--scope user (default) or --scope workspace; --path is the subfolder of the repo that holds the skill; --consent skips the security confirmation prompt.
$ npx skills add microsoft/bocpy --skill review-loop -a gemini-cliProject install goes to .agents/skills/; add -g for ~/.gemini/skills/.
$ gh skill install microsoft/bocpy review-loop --agent gemini-cliProject scope by default (.agents/skills/); add --scope user for a personal install.
$ git clone --depth 1 https://github.com/microsoft/bocpy.git skills-src && mkdir -p .gemini/skills && cp -r skills-src/.github/skills/review-loop .gemini/skills/review-loop && rm -rf skills-srcUse ~/.gemini/skills/ instead of .gemini/skills for a personal install, then run /skills reload.
Gemini CLI skills documentation · loads skills from .gemini/skills/, .agents/skills/
Install the "review-loop" agent skill from https://github.com/microsoft/bocpy/tree/main/.github/skills/review-loop into .gemini/skills/review-loop/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "review-loop", then confirm the skill loads.Gemini CLI copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.
$ gh skill install microsoft/bocpy review-loopInstalls for Copilot at project scope by default; add --scope user for a personal install. Preview a skill first with gh skill preview. Needs GitHub CLI 2.90.0 or later (public preview).
$ npx skills add microsoft/bocpy --skill review-loop -a github-copilotProject install goes to .agents/skills/; add -g for ~/.copilot/skills/.
$ git clone --depth 1 https://github.com/microsoft/bocpy.git skills-src && mkdir -p .github/skills && cp -r skills-src/.github/skills/review-loop .github/skills/review-loop && rm -rf skills-srcUse ~/.copilot/skills/ instead of .github/skills for a personal install. Commit .github/skills so cloud agent and code review can use it.
GitHub Copilot skills documentation · loads skills from .github/skills/, .claude/skills/, .agents/skills/
Install the "review-loop" agent skill from https://github.com/microsoft/bocpy/tree/main/.github/skills/review-loop into .github/skills/review-loop/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "review-loop", then confirm the skill loads.GitHub Copilot copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.
$ npx skills add microsoft/bocpy --skill review-loop -a opencodeOpenCode documents no install command of its own. Project install goes to .agents/skills/; add -g for ~/.config/opencode/skills/.
$ gh skill install microsoft/bocpy review-loop --agent opencodeProject scope by default (.agents/skills/); add --scope user for a personal install.
$ git clone --depth 1 https://github.com/microsoft/bocpy.git skills-src && mkdir -p .opencode/skills && cp -r skills-src/.github/skills/review-loop .opencode/skills/review-loop && rm -rf skills-srcUse ~/.config/opencode/skills/ instead of .opencode/skills for a personal install.
OpenCode skills documentation · loads skills from .opencode/skills/, .claude/skills/, .agents/skills/
Install the "review-loop" agent skill from https://github.com/microsoft/bocpy/tree/main/.github/skills/review-loop into .opencode/skills/review-loop/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "review-loop", then confirm the skill loads.OpenCode copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.
Facts
- Skill name
review-loop- GitHub stars
- 201
- Token cost
- ~1.2k tokens
- SKILL.md length
- 639 words
- Files
- 1
- Skills in repo
- 9
- Repo updated
- First seen
- Licence
- MIT
At a glance
Iterative review loop with subagent reviewer. An agent skill from microsoft/bocpy.
- Works in 5 steps: Identify the Review Target → Spawn Reviewer Subagent → Present Findings → …
- : reviewing code changes
- SKILL.md covers When to Use, Severity Levels, Procedure and Guidelines
- Instructions only: no scripts, shell commands, URLs or credentials in SKILL.md
What it does
Review Loop is an agent skill from microsoft/bocpy, published by the product's own GitHub organization. Iterative review loop with subagent reviewer. Use when: reviewing code changes, reviewing plans, auditing implementations, validating fixes, running code review, performing quality checks, or when /review is invoked. Spawns a fresh subagent to review a user-specified target, reports severity-tagged findings, applies approved fixes, and repeats until the user is satisfied.
Its SKILL.md is about 1.2k tokens, which your agent loads only when the skill is triggered. It is a single SKILL.md file with no bundled scripts.
It sits in Development, covering Subagents and Code review. The repository describes itself as: Behavior-Oriented Concurrency in Python. The licence is MIT.
When your agent uses it
- : reviewing code changes
- Reviewing plans
- Auditing implementations
- Validating fixes
Example prompts
- “/review-loop”
Workflow steps
5 steps, taken from the step headings in SKILL.md.
What it can do on your machine
Read from SKILL.md and the folder at commit c8f3ceb. 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
Review Loop loads about 1.2k tokens when it runs. Until then it costs about 97 tokens; SKILL.md has 639 words of instructions outside code blocks.
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
The full file from microsoft/bocpy at commit c8f3ceb, republished under its MIT licence (© microsoft). 639 words, ~1,211 tokens.
Download SKILL.mdSave it as
.claude/skills/review-loop/SKILL.md (or your agent's skills folder).- name
- review-loop
- description
- Iterative review loop with subagent reviewer. Use when: reviewing code changes, reviewing plans, auditing implementations, validating fixes, running code review, performing quality checks, or when /review is invoked. Spawns a fresh subagent to review a user-specified target, reports severity-tagged findings, applies approved fixes, and repeats until the user is satisfied.
- argument-hint
- Describe or reference the target to review (file paths, plan, diff, etc.)
Review Loop
Iteratively review a target (code or plan) using a fresh subagent reviewer, present findings, apply fixes, and repeat until the user approves.
When to Use
- After implementing a non-trivial change
- Validating a plan before execution
- Auditing code for correctness, style, or security issues
- Any time you want an independent second opinion on work in progress
Severity Levels
Findings are tagged with one of four severities:
| Severity | Meaning |
|---|---|
| critical | Correctness bug, security vulnerability, or data loss risk. Must fix. |
| high | Likely bug, race condition, or significant design flaw. Should fix. |
| medium | Code smell, unclear logic, missing edge case, or maintainability concern. Recommended fix. |
| low | Style nit, naming suggestion, minor improvement. Fix at discretion. |
Procedure
1. Identify the Review Target
Ask the user what to review if not already specified. The target can be:
- One or more source files (by path)
- A plan (in session memory or conversation)
- A diff or set of changes
- A specific function, class, or module
Gather the full content of the target so it can be passed to the reviewer.
2. Spawn Reviewer Subagent
If a specific lens is requested (e.g., correctness-lens, security-lens,
adversarial-lens, etc.), use that named lens agent operating in review
mode. Otherwise, use a generic reviewer.
Launch a subagent with the following prompt structure:
You are a code reviewer performing a thorough review of the following target. Your job is to find bugs, design flaws, security issues, and quality problems.
Review target: {include the full content of the target here}
Additional context: {include relevant surrounding code, tests, or specifications the reviewer needs to understand the target}
Instructions:
Review the target for correctness, security, performance, readability, and adherence to project conventions.
For each issue found, report it in this exact format:
[SEVERITY] Short title
- Location: file path and line number (or plan step)
- Problem: what is wrong and why it matters
- Suggestion: concrete fix or remediation
where SEVERITY is one of: critical, high, medium, low.
If you find no issues, state explicitly that the target looks correct.
Do NOT fabricate issues. Only report genuine problems.
Order findings by severity (critical first).
Use the Explore subagent for read-only review of code. If the reviewer needs
to run tests or execute code to verify a finding, note that as an unverified
finding and let the main agent handle verification.
Show full SKILL.md (255 more words)Show less
3. Present Findings
After the reviewer returns:
- List all findings grouped by severity.
- For each finding, include the reviewer's suggested remediation.
- If the reviewer found no issues, report that explicitly.
- Ask the user which findings to address. The user may:
- Approve all fixes
- Select specific findings to fix
- Dismiss findings they disagree with
- Ask for clarification on any finding
4. Apply Fixes
For each approved finding:
- Implement the suggested fix (or an alternative if the user provides one).
- Briefly confirm each fix as it is applied.
If a fix is non-trivial or ambiguous, ask the user for guidance rather than guessing.
5. Check Exit Condition
After fixes are applied, ask the user:
All approved fixes have been applied. Should I run another review pass, or are we done?
- If the user wants another pass → go to step 2 with the updated target.
- If the user is satisfied → exit the loop.
Guidelines
- Fresh context per pass. Each reviewer subagent starts with no memory of previous passes. This prevents anchoring bias and ensures new eyes on the updated code.
- Do not auto-fix without approval. Always present findings and wait for the user to decide which to address.
- Verify critical findings. If the reviewer flags a critical or high issue, attempt to verify it (e.g., by running tests or tracing the code) before presenting it to the user. Mark unverified findings as such.
- Keep the loop bounded. If the reviewer returns only low-severity findings on two consecutive passes, suggest exiting the loop.
© microsoft, MIT. Rendered from Markdown: HTML in the file is shown as text, images as links, and headings moved down two levels. Raw file
Files
Just SKILL.md in .github/skills/review-loop of microsoft/bocpy.
Open the folder on GitHubat commit c8f3ceb
Compare with similar skills
Review Loop 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.
| Skill | Stars | Used in | Tokens | Auto-check | Licence | Repo updated |
|---|---|---|---|---|---|---|
| Review Loop this skillmicrosoft/bocpy | 201 | — | ~1.2k | Automated safety check: Pass | MIT | |
| GitHub Review Iterationprisma/orm | 48k | — | ~2.2k | Automated safety check: Pass | Apache-2.0 | |
| Cherry Studio PR ReviewCherryHQ/cherry-studio | 52k | — | ~3.9k | Automated safety check: Pass | AGPL-3.0 | |
| PR Reviewjaemk/self_update | 961 | — | ~1.5k | Automated safety check: Notes | MIT | |
| Cursor Composer Task DelegateChachamaru127/claude-code-harness | 3.2k | — | ~4.4k | Automated safety check: Notes | MIT | |
| Local PR Reviewwindmill-labs/windmill | 18k | — | ~995 | Automated safety check: Pass | Custom licence |
Similar skills
- Official
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.
PR Review
jaemk/self_update
Targeted, read-only review of a PR or checked-out branch. An agent skill from jaemk/self_update.
Cursor Composer Task Delegate
Chachamaru127/claude-code-harness
Hands one implementation task to Cursor Composer in an isolated git worktree, then reviews its diff and cherry-picks the result into the main branch.
Local PR Review
windmill-labs/windmill
Runs the same code review locally that GitHub's auto-review actions run on a PR, delegating to a fresh-context subagent so the review isn't biased by the main session's own reasoning.
Auto Devflow
HuangPuStar/FenixAgent
A skill your agent uses when starting an issue, bugfix, feature, or refactor that should be driven by multiple coordinated subagents: explore → plan → code → review, with the main agent acting as…
More from microsoft/bocpy
All 9 skills in this repo- Official
Branch Review
microsoft/bocpy
Multi-perspective code review for a branch before merging. An agent skill from microsoft/bocpy.
- Official
C Extensions With Bocpy
microsoft/bocpy
Write a C extension whose custom types can live inside a bocpy Cown and travel between worker sub-interpreters.
- Official
Commenting C And Python
microsoft/bocpy
Follow bocpy commenting and documentation conventions. An agent skill from microsoft/bocpy.
- Official
Finalize PR
microsoft/bocpy
Finalize a feature branch for merge. An agent skill from microsoft/bocpy.
- Official
Multi Perspective Plan
microsoft/bocpy
Multi-perspective planning with rebuttal rounds and adversarial review loop.
- Official
Testing Message Queue
microsoft/bocpy
Write tests for the bocpy message queue — the lock-free tag-based MPSC ring buffer.
Related
Categories
Questions about Review Loop
What does Review Loop do?
Iterative review loop with subagent reviewer. An agent skill from microsoft/bocpy. Review Loop is an agent skill from microsoft/bocpy, published by the product's own GitHub organization. Iterative review loop with subagent reviewer.
When should I use Review Loop?
Review Loop fits situations like: : reviewing code changes; reviewing plans; auditing implementations; validating fixes.
How do I install Review Loop in Claude Code?
Run `npx skills add microsoft/bocpy --skill review-loop -a claude-code`. Or copy the skill folder (.github/skills/review-loop in microsoft/bocpy) into .claude/skills/review-loop in your project. Claude Code loads it when a task matches its description.
How do I install Review Loop in Codex?
Run `npx skills add microsoft/bocpy --skill review-loop -a codex`. Or copy the skill folder (.github/skills/review-loop in microsoft/bocpy) into .agents/skills/review-loop in your project. Codex loads it when a task matches its description.
Can I use Review Loop 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 microsoft/bocpy --skill review-loop -a cursor` (or -a gemini-cli, github-copilot or opencode for the others). To copy it by hand, put the folder in .cursor/skills/review-loop, .gemini/skills/review-loop, .github/skills/review-loop and .opencode/skills/review-loop in your project.
What does Review Loop need to run?
SKILL.md names no scripts, command-line tools or credentials: Review Loop is instructions for the agent only.
Does Review Loop 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 Review Loop 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 Review Loop use?
Review Loop is published under the MIT licence (the repository's licence). It allows redistribution, so the full SKILL.md is shown on this page.
How many tokens does Review Loop use?
About 1.2k tokens (SKILL.md is roughly 4.8k characters). Agents keep only the skill's name and description in context until a task matches; then they load SKILL.md in full.
What are the alternatives to Review Loop?
Skills that share tags, products or a category with Review Loop: GitHub Review Iteration (prisma/orm, 48k stars), Cherry Studio PR Review (CherryHQ/cherry-studio, 52k stars), PR Review (jaemk/self_update, 961 stars) and Cursor Composer Task Delegate (Chachamaru127/claude-code-harness, 3.2k stars). The comparison table on this page puts their stars, adoption, token cost, safety result and licence side by side.
Who maintains Review Loop?
microsoft (a GitHub organization, an official publisher) maintains it in microsoft/bocpy, which has 201 GitHub stars. The repository holds 9 skills in this directory. The repository was last updated on September 28, 2026.
Source: microsoft/bocpy on GitHub. Facts on this page come from the repository at the commit we read; the author's words are quoted as theirs.