[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"project-78669":3},{"id":4,"name":5,"fullName":6,"owner":7,"repo":5,"description":8,"homepage":9,"htmlUrl":10,"language":11,"languages":10,"totalLinesOfCode":10,"stars":12,"forks":13,"watchers":14,"openIssues":15,"contributorsCount":16,"subscribersCount":16,"size":16,"stars1d":17,"stars7d":18,"stars30d":19,"stars90d":16,"forks30d":16,"starsTrendScore":20,"compositeScore":21,"rankGlobal":10,"rankLanguage":10,"license":22,"archived":23,"fork":23,"defaultBranch":24,"hasWiki":23,"hasPages":23,"topics":25,"createdAt":10,"pushedAt":10,"updatedAt":26,"readmeContent":27,"aiSummary":28,"trendingCount":16,"starSnapshotCount":16,"syncStatus":29,"lastSyncTime":30,"discoverSource":31},78669,"roborev","kenn-io\u002Froborev","kenn-io","Continuous background code review database for agents, work faster and smarter with accountability for every line of generated code.","https:\u002F\u002Froborev.io",null,"Go",1374,122,9,60,0,52,77,188,156,19.27,"MIT License",false,"main",[],"2026-06-12 02:03:48","\u003Cpicture>\n  \u003Csource media=\"(prefers-color-scheme: dark)\" srcset=\"https:\u002F\u002Fraw.githubusercontent.com\u002Froborev-dev\u002Froborev-docs\u002Fmain\u002Fpublic\u002Flogo-with-text-dark-bg.svg\">\n  \u003Cimg alt=\"roborev\" src=\"https:\u002F\u002Fraw.githubusercontent.com\u002Froborev-dev\u002Froborev-docs\u002Fmain\u002Fpublic\u002Flogo-with-text-light.svg\">\n\u003C\u002Fpicture>\n\n[![Go](https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FGo-1.21+-00ADD8?logo=go&logoColor=white)](https:\u002F\u002Fgo.dev\u002F)\n[![License: MIT](https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FLicense-MIT-yellow.svg)](https:\u002F\u002Fopensource.org\u002Flicenses\u002FMIT)\n[![Docs](https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FDocs-roborev.io-blue)](https:\u002F\u002Froborev.io)\n\n**[Documentation](https:\u002F\u002Froborev.io)** | **[Quick Start](https:\u002F\u002Froborev.io\u002Fquickstart\u002F)** | **[Installation](https:\u002F\u002Froborev.io\u002Finstallation\u002F)**\n\nContinuous code review for AI coding agents. roborev runs in the\nbackground, reviews every commit as agents write code, and surfaces\nissues in seconds -- before they compound. Pull code reviews into\nyour agentic loop while context is fresh.\n\n![roborev TUI](https:\u002F\u002Fraw.githubusercontent.com\u002Froborev-dev\u002Froborev-docs\u002Fmain\u002Fpublic\u002Ftui-hero.svg)\n\n## How It Works\n\n1. Run `roborev init` to install a post-commit hook\n2. Every commit triggers a background review -- agents write, roborev reads\n3. View findings in the TUI, feed them to your agent, or let `roborev fix` handle it\n\n## Quick Start\n\n```bash\ncd your-repo\nroborev init          # Install post-commit hook\ngit commit -m \"...\"   # Reviews happen automatically\nroborev tui           # View reviews in interactive UI\n```\n\n![roborev review](https:\u002F\u002Fraw.githubusercontent.com\u002Froborev-dev\u002Froborev-docs\u002Fmain\u002Fpublic\u002Ftui-review.svg)\n\n## Features\n\n- **Background Reviews** - Every commit is reviewed automatically via\n  git hooks. No workflow changes required.\n- **Auto-Fix** - `roborev fix` feeds review findings to an agent that\n  applies fixes and commits. `roborev refine` iterates until reviews pass.\n- **Code Analysis** - Built-in analysis types (duplication, complexity,\n  refactoring, test fixtures, dead code, security) that agents can fix\n  automatically.\n- **Multi-Agent** - Works with Codex, Claude Code, Gemini, Copilot,\n  OpenCode, Cursor, Kiro, Kilo, Droid, and Pi.\n- **Runs Locally** - No hosted service or additional infrastructure.\n  Reviews are orchestrated on your machine using the coding agents\n  you already have configured.\n- **Interactive TUI** - Real-time review queue with vim-style navigation.\n- **Review Verification** - `roborev compact` verifies findings against\n  current code, filters false positives, and consolidates related issues\n  into a single review.\n- **Extensible Hooks** - Run shell commands on review events. Built-in\n  [beads](https:\u002F\u002Fgithub.com\u002Fsteveyegge\u002Fbeads) integration creates trackable issues from\n  review failures automatically.\n\n## The Agentic Fix Loop\n\nWhen reviews find issues, copy-and-paste the reviews into your\ninteractive agent sessions, or invoke the `roborev:fix` skills. You\ncan also address open reviews on the command line non-interactively\nwith `roborev fix`.\n\n`roborev fix` shows the review findings to an agent, which applies\nchanges and commits. The new commit gets reviewed automatically,\nclosing the loop.\n\nFor fully automated iteration (advanced feature), use `refine`:\n\n```bash\nroborev refine                  # Fix, re-review, repeat until passing\n```\n\n`refine` runs in an isolated worktree and loops: fix findings, wait for\nre-review, fix again, until all reviews pass or `--max-iterations` is hit.\n\n## Code Analysis\n\nRun targeted analysis across your codebase and optionally auto-fix:\n\n```bash\nroborev analyze duplication .\u002F...           # Find duplication\nroborev analyze refactor --fix *.go         # Suggest and apply refactors\nroborev analyze complexity --wait main.go   # Analyze and show results\nroborev analyze test-fixtures *_test.go     # Find test helper opportunities\nroborev analyze security .\u002F...              # Find security risks in existing code\n```\n\nAvailable types: `test-fixtures`, `duplication`, `refactor`, `complexity`,\n`api-design`, `dead-code`, `architecture`, `security`.\n\nAnalysis jobs appear in the review queue. Use `roborev fix \u003Cid>` to\napply findings later, or pass `--fix` to apply immediately.\n\n## Installation\n\n**Shell Script (macOS \u002F Linux):**\n```bash\ncurl -fsSL https:\u002F\u002Froborev.io\u002Finstall.sh | bash\n```\n\n**Homebrew (macOS \u002F Linux):**\n```bash\nbrew install roborev-dev\u002Ftap\u002Froborev\n```\n\n**Windows (PowerShell):**\n```powershell\npowershell -ExecutionPolicy ByPass -c \"irm https:\u002F\u002Froborev.io\u002Finstall.ps1 | iex\"\n```\n\n**With Go:**\n```bash\ngo install go.kenn.io\u002Froborev\u002Fcmd\u002Froborev@latest\n```\n\n## Developer Setup\n\nThis repo uses [`prek`](https:\u002F\u002Fprek.j178.dev\u002F) for local pre-commit checks.\nThe hook is a local system hook that runs `make lint`, so pre-commit can apply\n`golangci-lint --fix` automatically instead of using the upstream\n`golangci-lint` pre-commit repository. The hook is configured with\n`always_run = true`, so it runs on every commit, not just commits that touch\nGo files.\n\n```bash\nbrew install prek     # or use your preferred prek install method\nprek install          # install the local git hook\nprek run --all-files  # run the configured checks manually\n```\n\nIf the hook rewrites files, re-stage them and re-run `git commit`. Use\n`make lint-ci` when you want a non-mutating lint check.\n\n## Commands\n\n| Command | Description |\n|---------|-------------|\n| `roborev init` | Initialize roborev in current repo |\n| `roborev tui` | Interactive terminal UI |\n| `roborev status` | Show daemon and queue status |\n| `roborev review \u003Csha>` | Queue a commit for review |\n| `roborev review --branch` | Review all commits on current branch |\n| `roborev review --dirty` | Review uncommitted changes |\n| `roborev fix` | Fix open reviews (or specify job IDs) |\n| `roborev refine` | Auto-fix loop: fix, re-review, repeat |\n| `roborev analyze \u003Ctype>` | Run code analysis with optional auto-fix |\n| `roborev compact` | Verify and consolidate open review findings |\n| `roborev show [sha]` | Display review for commit |\n| `roborev run \"\u003Ctask>\"` | Execute a task with an AI agent |\n| `roborev close \u003Cid>` | Close a review |\n| `roborev skills install` | Install agent skills for Claude\u002FCodex |\n\nSee [full command reference](https:\u002F\u002Froborev.io\u002Fcommands\u002F) for all options.\n\n## Configuration\n\nCreate `.roborev.toml` in your repo:\n\n```toml\nagent = \"claude-code\"\nsnapshot_dir = \".roborev\"\nreview_guidelines = \"\"\"\nProject-specific review instructions here.\n\"\"\"\n```\n\n`snapshot_dir` must be repo-relative. `roborev init` ensures it is ignored in `.gitignore`; snapshot creation also adds a local `.git\u002Finfo\u002Fexclude` fallback for existing checkouts whose ignore setup is stale.\n\nSee [configuration guide](https:\u002F\u002Froborev.io\u002Fconfiguration\u002F) for all options.\n\n### Environment Variables\n\n| Variable | Description |\n|----------|-------------|\n| `ROBOREV_DATA_DIR` | Override default data directory (`~\u002F.roborev`) |\n| `ROBOREV_COLOR_MODE` | TUI color theme: `auto` (default), `dark`, `light`, `none` |\n| `NO_COLOR` | Set to any value to disable all color output ([no-color.org](https:\u002F\u002Fno-color.org)) |\n\n## Supported Agents\n\n| Agent | Install |\n|-------|---------|\n| Codex | `npm install -g @openai\u002Fcodex` |\n| Claude Code | `npm install -g @anthropic-ai\u002Fclaude-code` |\n| Gemini | `curl -fsSL https:\u002F\u002Fantigravity.google\u002Fcli\u002Finstall.sh \\| bash` (preferred Antigravity CLI) or `npm install -g @google\u002Fgemini-cli` |\n| Copilot | `npm install -g @github\u002Fcopilot` |\n| OpenCode | `go install github.com\u002Fopencode-ai\u002Fopencode@latest` |\n| Cursor | [cursor.com](https:\u002F\u002Fwww.cursor.com\u002F) |\n| Kiro | [kiro.dev](https:\u002F\u002Fkiro.dev\u002F) |\n| Kilo | `npm install -g @kilocode\u002Fcli` |\n| Droid | [factory.ai](https:\u002F\u002Ffactory.ai\u002F) |\n| Pi | [pi.dev](https:\u002F\u002Fpi.dev\u002F) |\n\nroborev auto-detects installed agents.\n\n### Routing Claude Code to a proxy (Ollama, LiteLLM, etc.)\n\nThe `claude-code` agent accepts a model spec of the form `\u003Cmodel>@\u003Cbase_url>`.\nWhen `\u003Cbase_url>` starts with `http(s):\u002F\u002F`, roborev points Claude Code at\nthat endpoint and pins all tier aliases (Opus\u002FSonnet\u002FHaiku\u002Fsubagent) to the\ngiven model.\n\n```toml\n# .roborev.toml — local Ollama for reviews, real Anthropic for fixes\nagent = \"claude-code\"\nreview_model = \"glm-5.1:cloud@http:\u002F\u002F127.0.0.1:11434\"\nfix_model    = \"sonnet\"\n```\n\nOr via CLI: `roborev review --model 'glm-5.1:cloud@http:\u002F\u002F127.0.0.1:11434'`.\n\n**Proxy auth.** Set `ROBOREV_CLAUDE_PROXY_TOKEN` to forward a bearer token\nto the proxy as `ANTHROPIC_AUTH_TOKEN`. If unset, roborev sends a placeholder\n(sufficient for gateways that don't check the header, such as Ollama).\nroborev does *not* forward `ANTHROPIC_API_KEY` to proxy endpoints — that\nwould leak a real Anthropic credential to arbitrary third parties.\n\n**URL restrictions.** Proxy URLs must not embed `user:pass@` credentials\n(use `ROBOREV_CLAUDE_PROXY_TOKEN`); `http:\u002F\u002F` is only accepted for loopback\nhosts (`127.0.0.1`, `::1`, `localhost`) so plaintext endpoints can't receive\ntokens over the wire. Use `https:\u002F\u002F` for remote proxies. The full URL\n(including any path or query string) is forwarded as-is to\n`ANTHROPIC_BASE_URL`, so include the path your gateway expects (e.g.\nLiteLLM may want a trailing `\u002Fv1`; Ollama wants no path).\n\n**Environment behavior (breaking change in this release).** When the\n`claude-code` agent runs, roborev always strips inherited `ANTHROPIC_API_KEY`,\n`ANTHROPIC_BASE_URL`, `ANTHROPIC_AUTH_TOKEN`,\n`ANTHROPIC_DEFAULT_{OPUS,SONNET,HAIKU}_MODEL`, and `CLAUDE_CODE_SUBAGENT_MODEL`\nfrom the child environment. If you were previously routing Claude Code by\nexporting these vars in your shell, switch to the `\u003Cmodel>@\u003Cbase_url>` spec\ninstead. For native (non-proxy) mode, configure `ANTHROPIC_API_KEY` via\nroborev's config (it is re-injected from roborev's stored key, not inherited\nfrom the operator's shell).\n\n## Security Model\n\nroborev delegates code review and fix tasks to AI coding agents that\nhave shell access. Review agents may execute read-only git and shell\ncommands to inspect diffs; fix agents run in isolated worktrees with\nfull tool access.\n\n**roborev is designed for use with trusted codebases.** The review\nprompt includes diff content and commit messages from the repository.\nIf you are reviewing untrusted code (e.g., open-source contributions\nfrom unknown authors), run roborev inside a sandboxed environment\n(container, VM, or similar) to limit the blast radius of any\nprompt-injection attack that could cause an agent to execute\nunintended commands.\n\n## Documentation\n\nFull documentation available at **[roborev.io](https:\u002F\u002Froborev.io)**:\n\n- [Quick Start](https:\u002F\u002Froborev.io\u002Fquickstart\u002F)\n- [Installation](https:\u002F\u002Froborev.io\u002Finstallation\u002F)\n- [Commands Reference](https:\u002F\u002Froborev.io\u002Fcommands\u002F)\n- [Configuration](https:\u002F\u002Froborev.io\u002Fconfiguration\u002F)\n- [Auto-Fixing with Refine](https:\u002F\u002Froborev.io\u002Fguides\u002Fauto-fixing\u002F)\n- [Code Analysis and Assisted Refactoring](https:\u002F\u002Froborev.io\u002Fguides\u002Fassisted-refactoring\u002F)\n- [Hooks](https:\u002F\u002Froborev.io\u002Fguides\u002Fhooks\u002F)\n- [Agent Skills](https:\u002F\u002Froborev.io\u002Fguides\u002Fagent-skills\u002F)\n- [PostgreSQL Sync](https:\u002F\u002Froborev.io\u002Fguides\u002Fpostgres-sync\u002F)\n\nFor local development in this repo, install hooks with `prek install` or run\n`make install-hooks` as a thin wrapper around `prek install`.\n\n## License\n\nMIT\n","roborev 是一个持续后台代码审查工具，专为AI编码代理设计，能够实时审查每一行生成的代码。其核心功能包括通过git钩子自动触发背景审查、自动修复发现的问题、内置多种代码分析类型（如重复代码、复杂度等）以及支持多款主流AI编码助手。特别适用于需要快速迭代且对代码质量有高要求的开发场景中，帮助开发者在保持高效工作的同时确保代码的可维护性和安全性。此外，roborev提供了交互式TUI界面，便于用户直观地查看和处理审查结果，并且整个过程完全本地运行，无需额外的云服务或基础设施支持。",2,"2026-06-11 03:57:07","high_star"]