{
  "schemaVersion": "1.0",
  "item": {
    "slug": "review-evo",
    "name": "ReviewEvo",
    "source": "tencent",
    "type": "skill",
    "category": "开发工具",
    "sourceUrl": "https://clawhub.ai/8co/review-evo",
    "canonicalUrl": "https://clawhub.ai/8co/review-evo",
    "targetPlatform": "OpenClaw"
  },
  "install": {
    "downloadMode": "redirect",
    "downloadUrl": "/downloads/review-evo",
    "sourceDownloadUrl": "https://wry-manatee-359.convex.site/api/v1/download?slug=review-evo",
    "sourcePlatform": "tencent",
    "targetPlatform": "OpenClaw",
    "installMethod": "Manual import",
    "extraction": "Extract archive",
    "prerequisites": [
      "OpenClaw"
    ],
    "packageFormat": "ZIP package",
    "includedAssets": [
      "SKILL.md"
    ],
    "primaryDoc": "SKILL.md",
    "quickSetup": [
      "Download the package from Yavira.",
      "Extract the archive and review SKILL.md first.",
      "Import or place the package into your OpenClaw setup."
    ],
    "agentAssist": {
      "summary": "Hand the extracted package to your coding agent with a concrete install brief instead of figuring it out manually.",
      "steps": [
        "Download the package from Yavira.",
        "Extract it into a folder your agent can access.",
        "Paste one of the prompts below and point your agent at the extracted folder."
      ],
      "prompts": [
        {
          "label": "New install",
          "body": "I downloaded a skill package from Yavira. Read SKILL.md from the extracted folder and install it by following the included instructions. Tell me what you changed and call out any manual steps you could not complete."
        },
        {
          "label": "Upgrade existing",
          "body": "I downloaded an updated skill package from Yavira. Read SKILL.md from the extracted folder, compare it with my current installation, and upgrade it while preserving any custom configuration unless the package docs explicitly say otherwise. Summarize what changed and any follow-up checks I should run."
        }
      ]
    },
    "sourceHealth": {
      "source": "tencent",
      "status": "healthy",
      "reason": "direct_download_ok",
      "recommendedAction": "download",
      "checkedAt": "2026-05-07T17:22:31.273Z",
      "expiresAt": "2026-05-14T17:22:31.273Z",
      "httpStatus": 200,
      "finalUrl": "https://wry-manatee-359.convex.site/api/v1/download?slug=afrexai-annual-report",
      "contentType": "application/zip",
      "probeMethod": "head",
      "details": {
        "probeUrl": "https://wry-manatee-359.convex.site/api/v1/download?slug=afrexai-annual-report",
        "contentDisposition": "attachment; filename=\"afrexai-annual-report-1.0.0.zip\"",
        "redirectLocation": null,
        "bodySnippet": null
      },
      "scope": "source",
      "summary": "Source download looks usable.",
      "detail": "Yavira can redirect you to the upstream package for this source.",
      "primaryActionLabel": "Download for OpenClaw",
      "primaryActionHref": "/downloads/review-evo"
    },
    "validation": {
      "installChecklist": [
        "Use the Yavira download entry.",
        "Review SKILL.md after the package is downloaded.",
        "Confirm the extracted package contains the expected setup assets."
      ],
      "postInstallChecks": [
        "Confirm the extracted package includes the expected docs or setup files.",
        "Validate the skill or prompts are available in your target agent workspace.",
        "Capture any manual follow-up steps the agent could not complete."
      ]
    },
    "downloadPageUrl": "https://openagent3.xyz/downloads/review-evo",
    "agentPageUrl": "https://openagent3.xyz/skills/review-evo/agent",
    "manifestUrl": "https://openagent3.xyz/skills/review-evo/agent.json",
    "briefUrl": "https://openagent3.xyz/skills/review-evo/agent.md"
  },
  "agentAssist": {
    "summary": "Hand the extracted package to your coding agent with a concrete install brief instead of figuring it out manually.",
    "steps": [
      "Download the package from Yavira.",
      "Extract it into a folder your agent can access.",
      "Paste one of the prompts below and point your agent at the extracted folder."
    ],
    "prompts": [
      {
        "label": "New install",
        "body": "I downloaded a skill package from Yavira. Read SKILL.md from the extracted folder and install it by following the included instructions. Tell me what you changed and call out any manual steps you could not complete."
      },
      {
        "label": "Upgrade existing",
        "body": "I downloaded an updated skill package from Yavira. Read SKILL.md from the extracted folder, compare it with my current installation, and upgrade it while preserving any custom configuration unless the package docs explicitly say otherwise. Summarize what changed and any follow-up checks I should run."
      }
    ]
  },
  "documentation": {
    "source": "clawhub",
    "primaryDoc": "SKILL.md",
    "sections": [
      {
        "title": "ReviewEvo",
        "body": "A self-improving code reviewer. It analyzes your git history, identifies risk hotspots, learns your team's conventions, and builds a persistent knowledge base that sharpens every review.\n\nNo external services. No API keys. No dependencies. It uses git and the agent's built-in tools — nothing else.\n\nFollow these steps in order. Complete each step fully before moving to the next."
      },
      {
        "title": "Prerequisites",
        "body": "Verify git is available:\n\nRun git --version and confirm output\n\nThe user must be inside a git repository with at least 20 commits of history. Run git rev-list --count HEAD to confirm. If fewer than 20 commits, warn the user that analysis will be limited but can still proceed."
      },
      {
        "title": "Step 1 — Detect Project and Load Prior Learnings",
        "body": "Check if this project has been reviewed before:\n\nls .review-evo/learnings.md 2>/dev/null\n\nIf the file exists: Read .review-evo/learnings.md in full. This contains findings from prior runs. Reference these throughout the review — confirm resolved issues, track recurring patterns, and build on previous analysis. Tell the user: \"I found learnings from a previous review. I'll build on those.\"\n\nIf the file does not exist: This is a first run. Tell the user you'll create the knowledge base after analysis.\n\nThen detect the project setup by checking for these files in the repo root:\n\ntsconfig.json → TypeScript\npackage.json → Node.js (read scripts for build/test/lint commands)\nrequirements.txt or pyproject.toml → Python\ngo.mod → Go\nCargo.toml → Rust\npom.xml or build.gradle → Java\n\nReport what you found and confirm with the user."
      },
      {
        "title": "Step 2 — Analyze Git History",
        "body": "Run each of these commands and capture the output. Do not summarize prematurely — collect all data before drawing conclusions.\n\nRecent activity (last 50 commits):\n\ngit log --oneline -50\n\nContributor breakdown:\n\ngit log --since=\"6 months ago\" --format=\"%an\" | sort | uniq -c | sort -rn\n\nHigh-churn files (most frequently modified):\n\ngit log --since=\"3 months ago\" --diff-filter=M --name-only --pretty=format: | sort | uniq -c | sort -rn | head -25\n\nLarge recent diffs (potential complexity bombs):\n\ngit log --since=\"1 month ago\" --pretty=format:\"%h %s\" --shortstat | head -60\n\nFiles with the most authors (knowledge-spread risk):\n\ngit log --since=\"6 months ago\" --pretty=format:\"%an\" --name-only | awk '/^$/{next} !author{author=$0;next} {files[author][$0]++; allfiles[$0]++} END{for(f in allfiles) {n=0; for(a in files) if(f in files[a]) n++; if(n>1) print n,f}}' | sort -rn | head -15\n\nIf the awk command fails on the platform, fall back to:\n\ngit log --since=\"6 months ago\" --format=\"%an\" --name-only | head -200\n\nand manually count distinct authors per file from the output."
      },
      {
        "title": "Step 3 — Build Review Profile",
        "body": "Using the data from Step 2, analyze and report on each of these dimensions:"
      },
      {
        "title": "Churn Hotspots",
        "body": "Files modified more than 5 times in 3 months are hotspots. For each one:\n\nRead the file\nAssess complexity (function count, nesting depth, line count)\nFlag if it lacks corresponding test coverage (look for matching *.test.*, *.spec.*, or test_* files)"
      },
      {
        "title": "Convention Patterns",
        "body": "From the recent commits and file contents, identify:\n\nNaming conventions (camelCase, snake_case, kebab-case for files)\nImport patterns (relative vs absolute, barrel files)\nError handling patterns (try/catch, Result types, error callbacks)\nComment density and style"
      },
      {
        "title": "Risk Indicators",
        "body": "Flag any of these:\n\nFiles over 400 lines with no tests\nFunctions over 50 lines\nTODOs or FIXMEs older than 30 days (check git log -1 --format=%cr on lines containing them)\nDependencies with known issues (check lock file age)\nSingle-author files in critical paths (bus factor risk)"
      },
      {
        "title": "Strengths",
        "body": "Also identify what the codebase does well:\n\nConsistent patterns\nGood test coverage areas\nClean separation of concerns\nRecent improvements visible in git history"
      },
      {
        "title": "Step 4 — Deliver the Review",
        "body": "Ask the user what they want reviewed:\n\nWhat would you like me to focus on?\n(a) Full codebase health report\n(b) A specific branch or PR diff (provide branch name)\n(c) Current working changes (git diff)\n(d) A specific file or directory"
      },
      {
        "title": "For option (a) — Full Health Report",
        "body": "Compile all findings from Step 3 into a structured report with sections: Hotspots, Risks, Conventions, Strengths, and Recommendations. Rank findings by severity (critical, warning, info)."
      },
      {
        "title": "For option (b) — Branch/PR Review",
        "body": "Run git diff main...{branch} (or the appropriate target branch). Analyze the diff through the lens of the patterns found in Step 3. Flag deviations from conventions, new risk introductions, and missing test coverage for changed code."
      },
      {
        "title": "For option (c) — Working Changes",
        "body": "Run git diff and git diff --cached. Apply the same analysis as option (b)."
      },
      {
        "title": "For option (d) — Targeted Review",
        "body": "Read the specified files. Analyze against the patterns and conventions discovered. Provide focused, actionable feedback.\n\nFor all options, structure each finding as:\n\nWhat: The specific issue or observation\nWhere: File and line range\nWhy it matters: Impact on maintainability, reliability, or security\nSuggestion: Concrete fix or improvement"
      },
      {
        "title": "Step 5 — Store Learnings",
        "body": "After delivering the review, persist findings for future runs.\n\nCreate the directory if it doesn't exist:\n\nmkdir -p .review-evo\n\nWrite (or append to) .review-evo/learnings.md with the following structure:\n\n## Review — {YYYY-MM-DD}\n\n### Project Profile\n- Language: {detected}\n- Key patterns: {conventions found}\n- Active contributors: {count}\n\n### Hotspots\n{list of high-churn files with context}\n\n### Recurring Patterns\n{patterns that appeared in this and prior reviews}\n\n### Resolved\n{items from prior reviews that are no longer flagged}\n\n### Open Risks\n{current findings ranked by severity}\n\nIf the file already exists, append the new review section. Do not overwrite prior entries — the history is the value.\n\nTell the user: \"Learnings saved. Next time I review this project, I'll build on these findings.\"\n\nAlso recommend adding .review-evo/ to the project's .gitignore if it's not already there — these are local analysis artifacts, not source code."
      },
      {
        "title": "Troubleshooting",
        "body": "\"Not a git repository\" — Run the skill from inside a git repo, or provide the path to one.\nawk command fails — Some platforms have limited awk. The skill includes fallback commands for each analysis step.\nVery large repos (10K+ commits) — The --since flags keep queries bounded. If commands are still slow, narrow the date range.\nMonorepo — Ask the user which subdirectory to focus on and scope all git commands with -- {path}."
      }
    ],
    "body": "ReviewEvo\n\nA self-improving code reviewer. It analyzes your git history, identifies risk hotspots, learns your team's conventions, and builds a persistent knowledge base that sharpens every review.\n\nNo external services. No API keys. No dependencies. It uses git and the agent's built-in tools — nothing else.\n\nFollow these steps in order. Complete each step fully before moving to the next.\n\nPrerequisites\n\nVerify git is available:\n\nRun git --version and confirm output\n\nThe user must be inside a git repository with at least 20 commits of history. Run git rev-list --count HEAD to confirm. If fewer than 20 commits, warn the user that analysis will be limited but can still proceed.\n\nStep 1 — Detect Project and Load Prior Learnings\n\nCheck if this project has been reviewed before:\n\nls .review-evo/learnings.md 2>/dev/null\n\n\nIf the file exists: Read .review-evo/learnings.md in full. This contains findings from prior runs. Reference these throughout the review — confirm resolved issues, track recurring patterns, and build on previous analysis. Tell the user: \"I found learnings from a previous review. I'll build on those.\"\n\nIf the file does not exist: This is a first run. Tell the user you'll create the knowledge base after analysis.\n\nThen detect the project setup by checking for these files in the repo root:\n\ntsconfig.json → TypeScript\npackage.json → Node.js (read scripts for build/test/lint commands)\nrequirements.txt or pyproject.toml → Python\ngo.mod → Go\nCargo.toml → Rust\npom.xml or build.gradle → Java\n\nReport what you found and confirm with the user.\n\nStep 2 — Analyze Git History\n\nRun each of these commands and capture the output. Do not summarize prematurely — collect all data before drawing conclusions.\n\nRecent activity (last 50 commits):\n\ngit log --oneline -50\n\n\nContributor breakdown:\n\ngit log --since=\"6 months ago\" --format=\"%an\" | sort | uniq -c | sort -rn\n\n\nHigh-churn files (most frequently modified):\n\ngit log --since=\"3 months ago\" --diff-filter=M --name-only --pretty=format: | sort | uniq -c | sort -rn | head -25\n\n\nLarge recent diffs (potential complexity bombs):\n\ngit log --since=\"1 month ago\" --pretty=format:\"%h %s\" --shortstat | head -60\n\n\nFiles with the most authors (knowledge-spread risk):\n\ngit log --since=\"6 months ago\" --pretty=format:\"%an\" --name-only | awk '/^$/{next} !author{author=$0;next} {files[author][$0]++; allfiles[$0]++} END{for(f in allfiles) {n=0; for(a in files) if(f in files[a]) n++; if(n>1) print n,f}}' | sort -rn | head -15\n\n\nIf the awk command fails on the platform, fall back to:\n\ngit log --since=\"6 months ago\" --format=\"%an\" --name-only | head -200\n\n\nand manually count distinct authors per file from the output.\n\nStep 3 — Build Review Profile\n\nUsing the data from Step 2, analyze and report on each of these dimensions:\n\nChurn Hotspots\n\nFiles modified more than 5 times in 3 months are hotspots. For each one:\n\nRead the file\nAssess complexity (function count, nesting depth, line count)\nFlag if it lacks corresponding test coverage (look for matching *.test.*, *.spec.*, or test_* files)\nConvention Patterns\n\nFrom the recent commits and file contents, identify:\n\nNaming conventions (camelCase, snake_case, kebab-case for files)\nImport patterns (relative vs absolute, barrel files)\nError handling patterns (try/catch, Result types, error callbacks)\nComment density and style\nRisk Indicators\n\nFlag any of these:\n\nFiles over 400 lines with no tests\nFunctions over 50 lines\nTODOs or FIXMEs older than 30 days (check git log -1 --format=%cr on lines containing them)\nDependencies with known issues (check lock file age)\nSingle-author files in critical paths (bus factor risk)\nStrengths\n\nAlso identify what the codebase does well:\n\nConsistent patterns\nGood test coverage areas\nClean separation of concerns\nRecent improvements visible in git history\nStep 4 — Deliver the Review\n\nAsk the user what they want reviewed:\n\nWhat would you like me to focus on? (a) Full codebase health report (b) A specific branch or PR diff (provide branch name) (c) Current working changes (git diff) (d) A specific file or directory\n\nFor option (a) — Full Health Report\n\nCompile all findings from Step 3 into a structured report with sections: Hotspots, Risks, Conventions, Strengths, and Recommendations. Rank findings by severity (critical, warning, info).\n\nFor option (b) — Branch/PR Review\n\nRun git diff main...{branch} (or the appropriate target branch). Analyze the diff through the lens of the patterns found in Step 3. Flag deviations from conventions, new risk introductions, and missing test coverage for changed code.\n\nFor option (c) — Working Changes\n\nRun git diff and git diff --cached. Apply the same analysis as option (b).\n\nFor option (d) — Targeted Review\n\nRead the specified files. Analyze against the patterns and conventions discovered. Provide focused, actionable feedback.\n\nFor all options, structure each finding as:\n\nWhat: The specific issue or observation\nWhere: File and line range\nWhy it matters: Impact on maintainability, reliability, or security\nSuggestion: Concrete fix or improvement\nStep 5 — Store Learnings\n\nAfter delivering the review, persist findings for future runs.\n\nCreate the directory if it doesn't exist:\n\nmkdir -p .review-evo\n\n\nWrite (or append to) .review-evo/learnings.md with the following structure:\n\n## Review — {YYYY-MM-DD}\n\n### Project Profile\n- Language: {detected}\n- Key patterns: {conventions found}\n- Active contributors: {count}\n\n### Hotspots\n{list of high-churn files with context}\n\n### Recurring Patterns\n{patterns that appeared in this and prior reviews}\n\n### Resolved\n{items from prior reviews that are no longer flagged}\n\n### Open Risks\n{current findings ranked by severity}\n\n\nIf the file already exists, append the new review section. Do not overwrite prior entries — the history is the value.\n\nTell the user: \"Learnings saved. Next time I review this project, I'll build on these findings.\"\n\nAlso recommend adding .review-evo/ to the project's .gitignore if it's not already there — these are local analysis artifacts, not source code.\n\nTroubleshooting\n\"Not a git repository\" — Run the skill from inside a git repo, or provide the path to one.\nawk command fails — Some platforms have limited awk. The skill includes fallback commands for each analysis step.\nVery large repos (10K+ commits) — The --since flags keep queries bounded. If commands are still slow, narrow the date range.\nMonorepo — Ask the user which subdirectory to focus on and scope all git commands with -- {path}."
  },
  "trust": {
    "sourceLabel": "tencent",
    "provenanceUrl": "https://clawhub.ai/8co/review-evo",
    "publisherUrl": "https://clawhub.ai/8co/review-evo",
    "owner": "8co",
    "version": "0.1.1",
    "license": null,
    "verificationStatus": "Indexed source record"
  },
  "links": {
    "detailUrl": "https://openagent3.xyz/skills/review-evo",
    "downloadUrl": "https://openagent3.xyz/downloads/review-evo",
    "agentUrl": "https://openagent3.xyz/skills/review-evo/agent",
    "manifestUrl": "https://openagent3.xyz/skills/review-evo/agent.json",
    "briefUrl": "https://openagent3.xyz/skills/review-evo/agent.md"
  }
}