ds-review-loop — Claude Code skill: what it does, how to install
ds-review-loop — a skill for Claude Code that runs an author and a reviewer in an iterative loop until they agree on the code changes.
MIT 548 installs ★ 320 updated 2026-10-28
npx skills add aicodeking/deepseek-stack --skill ds-review-loop -a claude-codeWhat ds-review-loop does
The skill iterates on code changes by passing them between the author and a fresh-context reviewer for up to four rounds. It stops when both agree on correctness or when a dispute requires user intervention.
The workflow tracks status via git commands, categorizes findings by severity, and logs outcomes in a structured ledger. Updates to pull requests include a summary of the rounds and the final verdict.
What it helps with
- review code in an iterative loop
- run a fresh-context reviewer on changes
- limit review rounds to four iterations
- log review findings and outcomes
Example prompts for the agent
- Run review loop on my current changes
- Get it reviewed before opening a pull request
- /ds-review-loop
How to install ds-review-loop
1Skills CLI (Vercel)
npx skills add aicodeking/deepseek-stack --skill ds-review-loop -a claude-code2Manually: put the files in place
~/.claude/skills/ds-review-loop/SKILL.mdCopy the whole skill folder; for one project use .claude/skills/ds-review-loop/
1Skills CLI (Vercel)
npx skills add aicodeking/deepseek-stack --skill ds-review-loop -a codex2Manually: put the files in place
.agents/skills/ds-review-loop/SKILL.mdFor all projects use ~/.agents/skills/
1Skills CLI (Vercel)
npx skills add aicodeking/deepseek-stack --skill ds-review-loop -a cursor2Manually: put the files in place
.cursor/skills/ds-review-loop/SKILL.mdCursor also reads .agents/skills/ and .claude/skills/
1GitHub CLI
gh skill install aicodeking/deepseek-stack skills/ds-review-loop2Manually: put the files in place
.github/skills/ds-review-loop/SKILL.md1Skills CLI (Vercel)
npx skills add aicodeking/deepseek-stack --skill ds-review-loop -a gemini-cli2Platform CLI
gemini skills install https://github.com/aicodeking/deepseek-stack --path skills/ds-review-loopManually: put the files in place
.devin/skills/ds-review-loop/SKILL.mdThe legacy .windsurf/skills/ path still works
Skills CLI (Vercel)
npx skills add aicodeking/deepseek-stack --skill ds-review-loop -a opencodeClaude.ai
Zip the folder and upload it in Settings → Capabilities → Skills
Step by step, with commands for each platform: How to install skills in Claude Code: five ways
Questions about ds-review-loop
How to install ds-review-loop in Claude Code?
Run npx skills add aicodeking/deepseek-stack --skill ds-review-loop -a claude-code in your terminal and reference ~/.claude/skills/ds-review-loop/SKILL.md.
What does ds-review-loop do?
It coordinates a multi-round review loop between an author and an independent reviewer until consensus is reached.
When does the review loop stop?
It stops when the reviewer approves, after four rounds, or if a disagreement remains unresolved after two exchanges.
Source text
SKILL.md · MIT
--- name: ds-review-loop description: "Runs an author and a fresh-context reviewer in a loop on the current change until both agree it is correct, for at most four rounds. Use before or after filing a PR, when the user says review loop, get it reviewed, or /ds-review-loop." --- Make a change more correct by letting one fresh reviewer push back until you both agree. ## Steps 1. Scope the change. Run `git status --short`. Do not stash, discard or include unrelated files. Find the base: `git fetch origin`, then `git merge-base HEAD origin/<base branch>`. Use one diff command for every round. 2. Write the goal and out-of-scope list in two sentences. Run the covering checks. Do not send code that fails them. 3. Pick the reviewer: a stronger model if the harness can reach one (subagent with a model setting, team worker, second CLI), or else a fresh DeepSeek context with no history. Record which one. 4. Send this brief: ``` Review only. Do not edit, commit, push, comment or start subagents. Repo: <path> Base: <sha> Diff: git diff <sha> [-- <paths>] Goal: <goal> Out of scope: <list> Checks: <results> Read the code around the diff. Find bugs, broken callers and gaps against the goal. Flag code the goal does not need. Ignore style unless it hides a bug. blocker = goal case is wrong. should = real bug users will hit. nit = anything else. Reply: VERDICT: APPROVE or CHANGES, then one line per finding: R1 [blocker|should|nit] file:line claim. Fix: suggestion ``` 5. Wait for the reply. Do not edit the scoped files while the review runs. A timeout or error is not approval. 6. Judge each finding yourself. Agree: make the smallest fix, preferring to delete or narrow code. Disagree: reply with evidence (file:line, test output). Nit or out of scope: decline in one line. 7. Keep a ledger: id, severity, outcome, evidence. Re-run checks and `git diff --shortstat`. If the diff grew, cut it back. 8. For the next round send the brief, the ledger and the changes since the last round. Ask the reviewer to check only open ids and changed code. ## Stop when - The reviewer returns APPROVE on the current diff, no blocker or should is open, and you have no concern left. - A round returns only nits. - Four rounds are done, or one item stays disputed after two exchanges. Report both sides in two or three lines each and let the user decide. This is not approval. ## Rules - One reviewer at a time. Never agree just to end the loop. - Do not push during the loop. Push afterward only if the user asked to push or ship. - If you update a PR, add: `Reviewed with ds-review-loop: <N> rounds, <reviewer>, verdict <verdict>.` - Report rounds, reviewer, fixed and declined findings with reasons, and diff size before and after.
© aicodeking · MIT · aicodeking/deepseek-stack