

Also from Kynth Studios


Also from Kynth Studios


Also from Kynth Studios
1# Code Review Agent23An LLM-powered code review agent built on the simple-agent SDK.45## Quick Start67```bash8# Install dependencies9npm install1011# Run code review12npm run start -- "帮我 review 当前的改动"13```1415## Usage Scenarios1617```bash18# Review uncommitted changes (staged + unstaged)19npm run start -- "帮我 review 当前的改动"2021# Review branch diff vs main22npm run start -- "帮我 review 当前 branch 新代码"2324# Review specific commit25npm run start -- "帮我 review commit abc123 的代码"2627# Review changes after a commit28npm run start -- "帮我 review commit abc123 之后的代码"2930# Review PR (requires gh CLI)31npm run start -- "帮我 review PR 42"32```3334## Environment Variables3536- `OPENAI_API_KEY` - Required for LLM API access37- `OPENAI_MODEL` - Model to use (default: gpt-5-codex)3839## Project Structure4041```42codereview-agent/43├── prompts/44│ └── system.md # System prompt (LLM behavior guide)45├── src/46│ ├── index.ts # Public exports47│ ├── agent.ts # createCodeReviewAgent factory48│ ├── cli.ts # CLI entry point49│ └── tools/50│ ├── bash.ts # Basic shell commands (read-only)51│ ├── read-file.ts # Read files in repo52│ ├── write-file.ts # Write review reports53│ ├── git.ts # Git commands (read-only)54│ └── gh.ts # GitHub CLI (read-only)55└── examples/56 └── basic-review.ts # Usage example57```5859## Development6061```bash62# Type check63npm run typecheck6465# Lint66npm run lint6768# Build69npm run build70```7172## Architecture7374The agent follows an LLM-driven architecture where:75761. **System Prompt** defines all business logic (intent understanding, tool selection, output format)772. **Agent Code** provides tools and security checks only783. **Tools** are read-only by design (git/gh commands are whitelisted)7980This design keeps the agent code minimal while allowing complex review workflows through prompt engineering.81
One repository carrying more than one format is the comparison this product exists for: does anyone actually write different content in each file, or is one a copy of the other?
| Repository | Format | Stack | Covers | Score | Changed |
|---|---|---|---|---|---|
| tyrchen/geektime-bootcamp-ai.cursor/rules/python-fastapi-backend.mdc · 230 | Cursor rules | setuptestlint-formatstyle+7 | 84/100 | 9 days ago | |
| tyrchen/geektime-bootcamp-ai.cursor/rules/rust-best-practices.mdc · 230 | Cursor rules | teststylearchdependencies+1 | 81/100 | 9 days ago | |
| tyrchen/geektime-bootcamp-ai.cursor/rules/specify-rules.mdc · 230 | Cursor rules | stylearch | 52/100 | 9 days ago | |
| tyrchen/geektime-bootcamp-aisite/CLAUDE.md · 230 | CLAUDE.md | agent-behaviour | 25/100 | 9 days ago | |
| tyrchen/geektime-bootcamp-aiw3/raflow/CLAUDE.md · 230 | CLAUDE.md | agent-behaviour | 25/100 | 9 days ago | |
| tyrchen/geektime-bootcamp-aiw6/opencode-introspection/CLAUDE.md · 230 | CLAUDE.md | setupbuildtestlint-format+10 | 84/100 | 9 days ago | |
| tyrchen/geektime-bootcamp-aiw6/opencode-introspection/visualizer/CLAUDE.md · 230 | CLAUDE.md | setupbuildtestlint-format+9 | 78/100 | 9 days ago | |
| tyrchen/geektime-bootcamp-aiw6/simple-agent/CLAUDE.md · 230 | CLAUDE.md | setupbuildtestlint-format+10 | 84/100 | 9 days ago | |
| tyrchen/geektime-bootcamp-aiw7/genslides/backend/CLAUDE.md · 230 | CLAUDE.md | testlint-formatstylearch+6 | 100/100 | 9 days ago | |
| tyrchen/geektime-bootcamp-aiw7/genslides/frontend/CLAUDE.md · 230 | CLAUDE.md | teststylearchtypes+5 | 88/100 | 9 days ago | |
| tyrchen/geektime-bootcamp-aiw5/pg-mcp/CLAUDE.md · 230 | CLAUDE.md | setuptestlint-formatstyle+5 | 86/100 | 9 days ago |
Same format, overlapping stack, ranked by quality.
| Repository | Format | Stack | Covers | Score | Changed |
|---|---|---|---|---|---|
| bagisto/bagistoCLAUDE.md · 28k | CLAUDE.md | setupbuildteststyle+5 | 100/100 | 7 days ago | |
| Adit-Jain-srm/NightmareNetCLAUDE.md · 46 | CLAUDE.md | buildtestlint-formatstyle+6 | 100/100 | 14 days ago | |
| tyrchen/geektime-bootcamp-aiw7/genslides/backend/CLAUDE.md · 230 | CLAUDE.md | testlint-formatstylearch+6 | 100/100 | 9 days ago | |
| nimbalyst/nimbalystpackages/android/CLAUDE.md · 1.5k | CLAUDE.md | setupbuildstylearch+2 | 100/100 | 14 days ago | |
| dotCMS/corecore-web/CLAUDE.md · 949 | CLAUDE.md | teststylearchtesting-strategy+3 | 100/100 | 14 days ago | |
| microsoft/playwrightCLAUDE.md · 95k | CLAUDE.md | buildtestlint-formatstyle+7 | 100/100 | 7 days ago | |
| tphakala/birdnet-goCLAUDE.md · 1.6k | CLAUDE.md | buildtestlint-formatstyle+8 | 100/100 | today | |
| dotCMS/coreCLAUDE.md · 949 | CLAUDE.md | setupbuildteststyle+7 | 99/100 | today |
A badge carrying the measured quality of the strongest agent config file in this repository, out of 100. It reads from this index every time somebody loads your page, so it changes when the measurement changes and there is nothing to keep up to date. Free, no account, and the value is not something you or we can set by hand.
[](https://rulestack.kynth.studio/configs/tyrchen-geektime-bootcamp-ai-w6-codereview-agent-claude)Would rather not hotlink us? Every badge is also served in shields.io’s endpoint schema, so shields renders the image and your readers never talk to our domain:
Published by Toolproof, the masthead over this index and eight others. The method behind the number is at toolproof.kynth.studio/methodology, and the whole thing is readable as JSON with no key at /api.