{"id":"0057640d-e9f5-4096-880b-5b329da01a9c","shortId":"HLBej2","kind":"skill","title":"autonomous-dev","tagline":"This skill should be used when the user wants to develop a feature, fix a bug, create a pull request, set up a git worktree, design a UI component, write test cases, push changes, check CI status, address review comments, resolve review threads, trigger bot reviews (/q review, /c","description":"# TDD Development Workflow\n\nA complete development workflow enforcing test-driven development, git worktree isolation, code review, CI verification, and E2E testing. Works in two modes: interactive (default) for human-guided sessions, and autonomous for fully unattended GitHub issue implementation.\n\n> **NON-NEGOTIABLE RULES -- Every step marked MANDATORY is required. You MUST NOT skip, defer, or ask the user whether to run these steps. Execute them automatically as part of the workflow. This includes: creating PRs, waiting for CI, running E2E tests, and addressing reviewer findings.**\n\n---\n\n## Mode Detection\n\n### Interactive Mode (default)\n\nUsed when a developer is present. The workflow:\n- Asks the user for design approval before proceeding to implementation\n- Presents design canvases and waits for feedback\n- Pauses at key decision points for user input\n- Reports final status and lets the user decide when to merge\n\n### Autonomous Mode\n\nTriggered when running inside the `scripts/autonomous-dev.sh` wrapper. The workflow:\n- Makes all decisions autonomously (see \"Decision Making Guidelines\" below)\n- Posts progress comments to the GitHub issue instead of asking questions\n- Creates design docs but skips interactive approval\n- Stops after verification -- does not merge (the review agent handles that)\n- Marks requirement checkboxes in the issue body as work progresses\n\n---\n\n## Cross-Platform Notes\n\n### Hooks Support\n\n| IDE/CLI | Hooks Support | Setup |\n|---------|--------------|-------|\n| Claude Code | Full | `hooks/README.md` |\n| Kiro CLI | Full | `hooks/README.md` |\n| Cursor | None | Follow steps manually |\n| Windsurf | None | Follow steps manually |\n| Gemini CLI | None | Follow steps manually |\n\n### Tool Name Mapping\n\nThis skill uses generic language. Map to your IDE's tools:\n- \"Execute in your terminal\" = Bash tool (Claude Code), terminal (Cursor), shell (Gemini CLI), etc.\n- \"Read the file\" = Read tool, file viewer, or `cat`\n- \"Create or edit the file\" = Write/Edit tool, editor, or manual editing\n- \"Use a subagent\" = Task/Agent tool (Claude Code), or follow the steps manually if unsupported\n- \"Load the skill\" = Skill tool (Claude Code), or read the referenced SKILL.md directly\n\n### Workflow Enforcement (Optional Hooks)\n\nIf your IDE/CLI supports hooks (Claude Code, Kiro CLI), install them from `hooks/` for hard enforcement. See `hooks/README.md` for setup.\n\nWithout hooks, follow each step manually -- the discipline is the same.\n\n---\n\n## Development Workflow Overview\n\nFollow this workflow for all feature development and bug fixes:\n\n```\nStep 1:  DESIGN CANVAS (Pencil MCP, if available)\nStep 2:  CREATE GIT WORKTREE (MANDATORY)\nStep 3:  WRITE TEST CASES (TDD)\nStep 4:  IMPLEMENT CHANGES\nStep 5:  LOCAL VERIFICATION\nStep 6:  CODE SIMPLIFICATION\nStep 7:  COMMIT AND CREATE PR          -- MANDATORY\nStep 8:  PR REVIEW AGENT               -- MANDATORY\nStep 9:  WAIT FOR ALL CI CHECKS        -- MANDATORY\nStep 10: ADDRESS REVIEWER BOT FINDINGS -- MANDATORY\nStep 11: ITERATE UNTIL NO FINDINGS\nStep 12: E2E TESTS & READY FOR MERGE   -- MANDATORY\nStep 13: CLEANUP WORKTREE\n```\n\n---\n\n## Step 1: Design Canvas\n\n**Available if your IDE has Pencil MCP.** If Pencil MCP is not available, create a design document (`docs/designs/<feature>.md`) manually instead.\n\n### When to Create a Design Canvas\n\n- New UI components or pages\n- Feature implementations with user-facing changes\n- Architecture decisions that benefit from visualization\n- Complex data flows or state management\n\n### Pencil MCP Workflow\n\n1. **Check editor state**: call `get_editor_state()` to see if a `.pen` file is open.\n2. **Open or create design file**: call `open_document(\"docs/designs/<feature>.pen\")` or `open_document(\"new\")`.\n3. **Get design guidelines** (if needed): call `get_guidelines(topic=\"landing-page|table|tailwind|code\")`.\n4. **Get style guide** for consistent design: call `get_style_guide_tags()` then `get_style_guide(tags=[...])`.\n5. **Create design elements**: call `batch_design(operations)` to create UI mockups, component hierarchy diagrams, data flow visualizations, and architecture diagrams.\n6. **Validate design visually**: call `get_screenshot()` to verify the design looks correct.\n7. **Document design decisions**: add text annotations explaining choices, component specifications, and interaction patterns.\n\n### Design Canvas Template Structure\n\n```\nFeature: <Feature Name>\nDate: YYYY-MM-DD\nStatus: Draft | In Review | Approved\n\n- UI Mockup / Wireframe\n- Component Architecture (component tree, props/state flow)\n- Data Flow Diagram (API calls, state management)\n- Design Notes (key decisions, accessibility, responsive behavior)\n```\n\n### Design Approval\n\n- **Interactive mode**: Present the design canvas to the user, get explicit approval, document feedback, update status to \"Approved.\"\n- **Autonomous mode**: Create the design doc and proceed immediately -- no approval gate.\n\n---\n\n## Step 2: Create Git Worktree (MANDATORY)\n\n**Every change MUST be developed in an isolated git worktree. Never develop directly on the main workspace.**\n\n> Enforced by `block-commit-outside-worktree.sh` hook (if hooks are installed). Commits outside worktrees are automatically blocked. Direct pushes to main are blocked by `block-push-to-main.sh`.\n\n### Why Worktrees?\n\n- **Isolation**: Each feature/fix gets its own directory, preventing cross-contamination\n- **Parallel work**: Multiple features can be in progress simultaneously\n- **Clean main workspace**: The main checkout stays on `main`, ready for quick checks\n- **Safe rollback**: Discard a worktree without affecting the main workspace\n\n### Worktree Creation Process\n\nExecute in your terminal:\n\n```bash\n# 1. Determine branch name based on change type\n#    feat/<name>, fix/<name>, refactor/<name>, etc.\nBRANCH_NAME=\"feat/my-feature\"\n\n# 2. Create worktree with new branch from main\ngit worktree add .worktrees/$BRANCH_NAME -b $BRANCH_NAME\n\n# 3. Enter the worktree\ncd .worktrees/$BRANCH_NAME\n\n# 4. Install dependencies (use your project's package manager)\nnpm install  # or: bun install, yarn install, pnpm install\n\n# 5. Verify clean baseline\nnpm run build && npm test\n```\n\n### Directory Convention\n\n| Item | Value |\n|------|-------|\n| Worktree root | `.worktrees/` (project-local, gitignored) |\n| Path pattern | `.worktrees/<branch-name>` |\n| Example | `.worktrees/feat/user-authentication` |\n\n### Safety Checks\n\nBefore creating any worktree, verify `.worktrees/` is in `.gitignore`:\n\n```bash\ngit check-ignore -q .worktrees 2>/dev/null || echo \"WARNING: .worktrees not in .gitignore!\"\n```\n\n### All Subsequent Steps Run INSIDE the Worktree\n\nAfter creating the worktree, **all development commands** (test, lint, build, commit, push) are executed from within the worktree directory. The main workspace is not touched until cleanup.\n\n---\n\n## Step 3: Write Test Cases (TDD)\n\nBefore writing any implementation code:\n\n1. Read the design canvas and requirements\n2. Identify all user scenarios, edge cases, and error handling paths\n3. Create or edit the test case document: `docs/test-cases/<feature>.md`\n   - List all test scenarios (happy path, edge cases, error handling)\n   - Assign test IDs (e.g., `TC-AUTH-001`)\n   - Define expected results and acceptance criteria\n4. Create unit test skeletons\n5. Create E2E test cases if applicable\n\n---\n\n## Step 4: Implement Changes\n\n- Write code following the test cases (inside the worktree)\n- Write new unit tests for new functionality\n- Update existing tests if behavior changed\n- Ensure implementation covers all test scenarios\n\n---\n\n## Step 5: Local Verification\n\nExecute in your terminal:\n\n```bash\nnpm run build\nnpm run test\n```\n\nFix any failures before proceeding. Deploy and verify locally if applicable.\n\n---\n\n## Step 6: Code Simplification\n\n1. Use a subagent if your IDE supports them (e.g., `code-simplifier:code-simplifier`), otherwise review the code manually for unnecessary complexity.\n2. Address simplification suggestions.\n3. Mark complete (if hooks are installed):\n   ```bash\n   hooks/state-manager.sh mark code-simplifier\n   ```\n\n---\n\n## Step 7: Commit and Create PR (MANDATORY)\n\n### Commit\n\nExecute in your terminal:\n\n```bash\ngit add <files>\ngit commit -m \"type(scope): description\"\ngit push -u origin <branch-name>\n```\n\n### Create PR\n\n```bash\ngh pr create --title \"type(scope): description\" --body \"$(cat <<'EOF'\n## Summary\n<1-3 bullet points describing the change>\n\n## Design\n- [ ] Design canvas created (`docs/designs/<feature>.pen`)\n- [ ] Design approved\n\n## Test Plan\n- [ ] Test cases documented (`docs/test-cases/<feature>.md`)\n- [ ] Build passes (`npm run build`)\n- [ ] Unit tests pass (`npm run test`)\n- [ ] CI checks pass\n- [ ] Code simplification review passed\n- [ ] PR review agent review passed\n- [ ] Reviewer bot findings addressed (no new findings)\n- [ ] E2E tests pass\n\n## Checklist\n- [ ] New unit tests written for new functionality\n- [ ] E2E test cases updated if needed\n- [ ] Documentation updated if needed\nEOF\n)\"\n```\n\n### Update PR Checklist\n\nAfter completing each step, update the PR description:\n\n```bash\ngh pr view {pr_number} --json body --jq '.body' > /tmp/pr_body.md\n# Edit the checklist (mark items as [x])\ngh pr edit {pr_number} --body \"$(cat /tmp/pr_body.md)\"\n```\n\n---\n\n## Step 8: PR Review Agent (MANDATORY)\n\n1. Use a subagent if your IDE supports them (e.g., `/pr-review-toolkit:review-pr`), otherwise perform a self-review against the PR diff.\n2. Address findings by severity:\n   - Critical/Severe: Must fix\n   - High: Must fix\n   - Medium: Should fix\n   - Low: Optional\n3. Mark complete (if hooks are installed):\n   ```bash\n   hooks/state-manager.sh mark pr-review\n   ```\n\n---\n\n## Step 9: Wait for All CI Checks (MANDATORY -- DO NOT SKIP)\n\nExecute in your terminal:\n\n```bash\n# Watch all checks until completion\ngh pr checks {pr_number} --watch --interval 30\n```\n\nALL checks must pass: Lint, Unit Tests, Build, Deploy Preview, E2E Tests.\n\nIf ANY check fails: analyze logs, fix, push, re-watch. DO NOT proceed until every check shows \"pass.\"\n\n### Checks to Monitor\n\n| Check | Description | Action if Failed |\n|-------|-------------|------------------|\n| CI / build-and-test | Build + unit tests | Fix code or update snapshots |\n| Security Scan | SAST, npm audit | Fix security issues |\n| Amazon Q Developer | Security review | Address findings, retrigger with `/q review` |\n| Codex | AI code review | Address findings, retrigger with `/codex review` |\n| Other review bots | Various checks | Address findings, retrigger per bot docs |\n\n---\n\n## Step 10: Address Reviewer Bot Findings (MANDATORY)\n\nMultiple review bots can provide automated code review findings on PRs:\n\n| Bot | Trigger Command | Bot Username |\n|-----|-----------------|------------|\n| Amazon Q Developer | `/q review` | `amazon-q-developer[bot]` |\n| Codex | `/codex review` | `codex[bot]` |\n| Other bots | See bot documentation | Varies |\n\n### Handling Bot Review Findings\n\n1. **Review all comments** -- read each finding carefully\n2. **Determine action**:\n   - Valid issue: fix the code and push\n   - False positive: reply explaining the design decision\n3. **Reply to each thread** -- use direct reply, not a general PR comment\n4. **Resolve each thread** -- mark conversation as resolved\n5. **Retrigger review** -- comment with the appropriate trigger (e.g., `/q review`, `/codex review`)\n\n### Retrigger Bot Reviews\n\n> **IMPORTANT:** Some bot reviewers (e.g., Amazon Q Developer) ignore `/q review` comments posted by GitHub App bot accounts. If your project uses `scripts/gh-as-user.sh`, you **MUST** use it to trigger bot reviews so the comment is attributed to a real user. Do NOT use the default `gh` wrapper for bot review triggers.\n\n```bash\n# Amazon Q Developer (use gh-as-user.sh to post as a real user)\nbash scripts/gh-as-user.sh pr comment {pr_number} --body \"/q review\"\n\n# Codex\nbash scripts/gh-as-user.sh pr comment {pr_number} --body \"/codex review\"\n```\n\nIf `scripts/gh-as-user.sh` is not available in your project, use `gh pr comment` directly as a fallback.\n\nWait 60-90 seconds for the review to complete, then check for new comments.\n\n---\n\n## Step 11: Iterate Until No Findings\n\n**Repeat until review bots find no more issues:**\n\n1. Address findings (fix code or explain design)\n2. Reply to each comment thread\n3. Resolve all threads\n4. Trigger review command (`/q review`, `/codex review`, etc.)\n5. Wait 60-90 seconds\n6. Check for new findings\n7. **If new findings: repeat from step 1**\n8. **Only proceed when no new positive findings appear**\n\n---\n\n## Step 12: E2E Tests & Ready for Merge (MANDATORY -- DO NOT SKIP)\n\n1. Run E2E tests against the deployed preview environment (all tests must pass; skipped agent-dependent tests are acceptable)\n2. Mark complete (if hooks are installed):\n   ```bash\n   hooks/state-manager.sh mark e2e-tests\n   ```\n3. Update PR checklist to show all items complete\n4. **STOP HERE**: report status to the user (interactive mode) or post a summary comment on the issue (autonomous mode)\n5. User or review agent decides when to merge\n\n---\n\n## Step 13: Cleanup Worktree\n\nAfter the PR is merged or closed, execute in your terminal:\n\n```bash\n# Return to main workspace\ncd $(git rev-parse --show-toplevel)\n\n# Remove the worktree\ngit worktree remove .worktrees/<branch-name>\n\n# Prune stale worktree references\ngit worktree prune\n```\n\n---\n\n## References\n\nFor detailed commands and conventions, consult:\n- **`references/commit-conventions.md`** -- Branch naming and commit message conventions\n- **`references/review-commands.md`** -- Complete `gh` CLI and GraphQL command reference\n- **`references/review-threads.md`** -- Review thread management, response patterns, and quick reference commands\n- **`references/autonomous-mode.md`** -- Decision making, resume awareness, requirement tracking, pre-existing changes, bot review integration, and error recovery (autonomous mode only)","tags":["autonomous","dev","team","zxkane","agent-skills","ai-agents","ai-code-review","autonomous-coding","autonomous-dev-team","ci-cd","claude-code","claude-code-hooks"],"capabilities":["skill","source-zxkane","skill-autonomous-dev","topic-agent-skills","topic-ai-agents","topic-ai-code-review","topic-autonomous-coding","topic-autonomous-dev-team","topic-ci-cd","topic-claude-code","topic-claude-code-hooks","topic-code-review-automation","topic-codex-cli","topic-coding-agents","topic-devops-automation"],"categories":["autonomous-dev-team"],"synonyms":[],"warnings":[],"endpointUrl":"https://skills.sh/zxkane/autonomous-dev-team/autonomous-dev","protocol":"skill","transport":"skills-sh","auth":{"type":"none","details":{"cli":"npx skills add zxkane/autonomous-dev-team","source_repo":"https://github.com/zxkane/autonomous-dev-team","install_from":"skills.sh"}},"qualityScore":"0.457","qualityRationale":"deterministic score 0.46 from registry signals: · indexed on github topic:agent-skills · 15 github stars · SKILL.md body (13,842 chars)","verified":false,"liveness":"unknown","lastLivenessCheck":null,"agentReviews":{"count":0,"score_avg":null,"cost_usd_avg":null,"success_rate":null,"latency_p50_ms":null,"narrative_summary":null,"summary_updated_at":null},"enrichmentModel":"deterministic:skill-github:v1","enrichmentVersion":1,"enrichedAt":"2026-04-22T13:03:20.329Z","embedding":null,"createdAt":"2026-04-19T00:40:45.072Z","updatedAt":"2026-04-22T13:03:20.329Z","lastSeenAt":"2026-04-22T13:03:20.329Z","tsv":"'-3':1182 '-90':1682,1738 '/c':52 '/codex':1459,1506,1577,1662,1732 '/dev/null':923 '/pr-review-toolkit':1308 '/q':50,1449,1498,1575,1591,1652,1730 '/tmp/pr_body.md':1276,1291 '001':1020 '1':406,484,541,821,975,1101,1181,1298,1520,1708,1752,1773 '10':459,1473 '11':466,1695 '12':472,1763 '13':480,1845 '2':414,557,724,836,922,982,1125,1322,1528,1716,1793 '3':420,572,853,965,993,1129,1338,1545,1722,1806 '30':1379 '4':426,588,861,1027,1040,1558,1726,1815 '5':430,605,879,1032,1072,1566,1735,1835 '6':434,626,1098,1740 '60':1681,1737 '7':438,639,1143,1745 '8':445,1293,1753 '9':451,1352 'accept':1025,1792 'access':688 'account':1599 'action':1416,1530 'add':643,846,1156 'address':41,137,460,1126,1229,1323,1445,1455,1466,1474,1709 'affect':809 'agent':235,448,1223,1296,1788,1839 'agent-depend':1787 'ai':1452 'amazon':1440,1495,1501,1587,1634 'amazon-q-develop':1500 'analyz':1396 'annot':645 'api':680 'app':1597 'appear':1761 'applic':1038,1096 'appropri':1572 'approv':158,226,667,692,704,710,721,1195 'architectur':526,624,672 'ask':110,153,218 'assign':1013 'attribut':1617 'audit':1436 'auth':1019 'autom':1484 'automat':120,758 'autonom':2,87,189,203,711,1833,1935 'autonomous-dev':1 'avail':412,487,499,1668 'awar':1922 'b':850 'base':825 'baselin':882 'bash':300,820,915,1079,1136,1154,1169,1266,1345,1366,1633,1645,1655,1800,1859 'batch':610 'behavior':690,1063 'benefit':529 'block':759,765 'block-commit-outside-worktree.sh':748 'block-push-to-main.sh':767 'bodi':244,1177,1273,1275,1289,1651,1661 'bot':48,462,1227,1463,1470,1476,1481,1490,1493,1504,1509,1511,1513,1517,1580,1584,1598,1611,1630,1703,1929 'branch':823,833,841,848,851,859,1894 'bug':19,403 'build':885,946,1082,1203,1207,1387,1421,1424 'build-and-test':1420 'bullet':1183 'bun':873 'call':545,563,578,595,609,630,681 'canva':408,486,513,654,698,979,1190 'canvas':165 'care':1527 'case':35,423,968,988,999,1010,1036,1048,1199,1246 'cat':318,1178,1290 'cd':857,1864 'chang':37,428,525,730,827,1042,1064,1187,1928 'check':38,456,542,802,905,918,1215,1357,1369,1374,1381,1394,1408,1411,1414,1465,1690,1741 'check-ignor':917 'checkbox':240 'checklist':1236,1257,1279,1809 'checkout':795 'choic':647 'ci':39,70,132,455,1214,1356,1419 'claud':258,302,335,349,366 'clean':790,881 'cleanup':481,963,1846 'cli':263,277,308,369,1903 'close':1854 'code':68,259,303,336,350,367,435,587,974,1044,1099,1112,1115,1120,1140,1217,1428,1453,1485,1535,1712 'code-simplifi':1111,1114,1139 'codex':1451,1505,1508,1654 'command':943,1492,1729,1889,1906,1917 'comment':43,211,1523,1557,1569,1593,1615,1648,1658,1675,1693,1720,1829 'commit':439,754,947,1144,1149,1158,1897 'complet':57,1131,1259,1340,1371,1688,1795,1814,1901 'complex':532,1124 'compon':32,516,617,648,671,673 'consist':593 'consult':1892 'contamin':780 'convent':889,1891,1899 'convers':1563 'correct':638 'cover':1067 'creat':20,128,220,319,415,441,500,510,560,606,614,713,725,837,907,938,994,1028,1033,1146,1167,1172,1191 'creation':814 'criteria':1026 'critical/severe':1327 'cross':249,779 'cross-contamin':778 'cross-platform':248 'cursor':266,305 'data':533,620,677 'date':658 'dd':662 'decid':185,1840 'decis':173,202,205,527,642,687,1544,1919 'default':80,144,1626 'defer':108 'defin':1021 'depend':863,1789 'deploy':1091,1388,1779 'describ':1185 'descript':1162,1176,1265,1415 'design':29,157,164,221,407,485,502,512,561,574,594,607,611,628,636,641,653,684,691,697,715,978,1188,1189,1194,1543,1715 'detail':1888 'detect':141 'determin':822,1529 'dev':3 'develop':14,54,58,64,148,392,401,733,740,942,1442,1497,1503,1589,1636 'diagram':619,625,679 'diff':1321 'direct':356,741,760,1551,1676 'directori':776,888,955 'discard':805 'disciplin':388 'doc':222,716,1471 'docs/designs':504,566,1192 'docs/test-cases':1001,1201 'document':503,565,570,640,705,1000,1200,1250,1514 'draft':664 'driven':63 'e.g':1016,1110,1307,1574,1586 'e2e':73,134,473,1034,1233,1244,1390,1764,1775,1804 'e2e-tests':1803 'echo':924 'edg':987,1009 'edit':321,329,996,1277,1286 'editor':326,543,547 'element':608 'enforc':60,358,376,746 'ensur':1065 'enter':854 'environ':1781 'eof':1179,1254 'error':990,1011,1933 'etc':309,832,1734 'everi':98,729,1407 'exampl':902 'execut':118,296,816,950,1075,1150,1362,1855 'exist':1060,1927 'expect':1022 'explain':646,1541,1714 'explicit':703 'face':524 'fail':1395,1418 'failur':1088 'fallback':1679 'fals':1538 'feat':829 'feat/my-feature':835 'featur':16,400,519,657,784 'feature/fix':772 'feedback':169,706 'file':312,315,323,554,562 'final':179 'find':139,463,470,1228,1232,1324,1446,1456,1467,1477,1487,1519,1526,1699,1704,1710,1744,1748,1760 'fix':17,404,830,1086,1329,1332,1335,1398,1427,1437,1533,1711 'flow':534,621,676,678 'follow':268,273,279,338,383,395,1045 'full':260,264 'fulli':89 'function':1058,1243 'gate':722 'gemini':276,307 'general':1555 'generic':288 'get':546,573,579,589,596,601,631,702,773 'gh':1170,1267,1284,1372,1627,1673,1902 'gh-as-user.sh':1638 'git':27,65,416,726,737,844,916,1155,1157,1163,1865,1875,1883 'github':91,214,1596 'gitignor':898,914,929 'graphql':1905 'guid':84,591,598,603 'guidelin':207,575,580 'handl':236,991,1012,1516 'happi':1007 'hard':375 'hierarchi':618 'high':1330 'hook':252,255,360,365,373,382,749,751,1133,1342,1797 'hooks/readme.md':261,265,378 'hooks/state-manager.sh':1137,1346,1801 'human':83 'human-guid':82 'id':1015 'ide':293,490,1107,1304 'ide/cli':254,363 'identifi':983 'ignor':919,1590 'immedi':719 'implement':93,162,427,520,973,1041,1066 'import':1582 'includ':127 'input':177 'insid':194,934,1049 'instal':370,753,862,871,874,876,878,1135,1344,1799 'instead':216,507 'integr':1931 'interact':79,142,225,651,693,1823 'interv':1378 'isol':67,736,770 'issu':92,215,243,1439,1532,1707,1832 'item':890,1281,1813 'iter':467,1696 'jq':1274 'json':1272 'key':172,686 'kiro':262,368 'land':583 'landing-pag':582 'languag':289 'let':182 'lint':945,1384 'list':1003 'load':344 'local':431,897,1073,1094 'log':1397 'look':637 'low':1336 'm':1159 'main':744,763,791,794,798,811,843,957,1862 'make':200,206,1920 'manag':537,683,869,1911 'mandatori':101,418,443,449,457,464,478,728,1148,1297,1358,1478,1769 'manual':270,275,281,328,341,386,506,1121 'map':284,290 'mark':100,238,1130,1138,1280,1339,1347,1562,1794,1802 'mcp':410,493,496,539 'md':505,1002,1202 'medium':1333 'merg':188,232,477,1768,1843,1852 'messag':1898 'mm':661 'mockup':616,669 'mode':78,140,143,190,694,712,1824,1834,1936 'monitor':1413 'multipl':783,1479 'must':105,731,1328,1331,1382,1606,1784 'name':283,824,834,849,852,860,1895 'need':577,1249,1253 'negoti':96 'never':739 'new':514,571,840,1053,1057,1231,1237,1242,1692,1743,1747,1758 'non':95 'non-negoti':94 'none':267,272,278 'note':251,685 'npm':870,883,886,1080,1083,1205,1211,1435 'number':1271,1288,1376,1650,1660 'open':556,558,564,569 'oper':612 'option':359,1337 'origin':1166 'otherwis':1117,1312 'outsid':755 'overview':394 'packag':868 'page':518,584 'parallel':781 'pars':1868 'part':122 'pass':1204,1210,1216,1220,1225,1235,1383,1410,1785 'path':899,992,1008 'pattern':652,900,1913 'paus':170 'pen':553,567,1193 'pencil':409,492,495,538 'per':1469 'perform':1313 'plan':1197 'platform':250 'pnpm':877 'point':174,1184 'posit':1539,1759 'post':209,1594,1640,1826 'pr':442,446,1147,1168,1171,1221,1256,1264,1268,1270,1285,1287,1294,1311,1320,1349,1373,1375,1556,1647,1649,1657,1659,1674,1808,1850 'pr-review':1348 'pre':1926 'pre-exist':1925 'present':150,163,695 'prevent':777 'preview':1389,1780 'proceed':160,718,1090,1405,1755 'process':815 'progress':210,247,788 'project':866,896,1602,1671 'project-loc':895 'props/state':675 'provid':1483 'prs':129,1489 'prune':1879,1885 'pull':22 'push':36,761,948,1164,1399,1537 'q':920,1441,1496,1502,1588,1635 'question':219 'quick':801,1915 're':1401 're-watch':1400 'read':310,313,352,976,1524 'readi':475,799,1766 'real':1620,1643 'recoveri':1934 'refactor':831 'refer':1882,1886,1907,1916 'referenc':354 'references/autonomous-mode.md':1918 'references/commit-conventions.md':1893 'references/review-commands.md':1900 'references/review-threads.md':1908 'remov':1872,1877 'repeat':1700,1749 'repli':1540,1546,1552,1717 'report':178,1818 'request':23 'requir':103,239,981,1923 'resolv':44,1559,1565,1723 'respons':689,1912 'result':1023 'resum':1921 'retrigg':1447,1457,1468,1567,1579 'return':1860 'rev':1867 'rev-pars':1866 'review':42,45,49,51,69,138,234,447,461,666,1118,1219,1222,1224,1226,1295,1310,1317,1350,1444,1450,1454,1460,1462,1475,1480,1486,1499,1507,1518,1521,1568,1576,1578,1581,1585,1592,1612,1631,1653,1663,1686,1702,1728,1731,1733,1838,1909,1930 'review-pr':1309 'rollback':804 'root':893 'rule':97 'run':115,133,193,884,933,1081,1084,1206,1212,1774 'safe':803 'safeti':904 'sast':1434 'scan':1433 'scenario':986,1006,1070 'scope':1161,1175 'screenshot':632 'scripts/autonomous-dev.sh':196 'scripts/gh-as-user.sh':1604,1646,1656,1665 'second':1683,1739 'secur':1432,1438,1443 'see':204,377,550,1512 'self':1316 'self-review':1315 'session':85 'set':24 'setup':257,380 'sever':1326 'shell':306 'show':1409,1811,1870 'show-toplevel':1869 'simplif':436,1100,1127,1218 'simplifi':1113,1116,1141 'simultan':789 'skeleton':1031 'skill':5,286,346,347 'skill-autonomous-dev' 'skill.md':355 'skip':107,224,1361,1772,1786 'snapshot':1431 'source-zxkane' 'specif':649 'stale':1880 'state':536,544,548,682 'status':40,180,663,708,1819 'stay':796 'step':99,117,269,274,280,340,385,405,413,419,425,429,433,437,444,450,458,465,471,479,483,723,932,964,1039,1071,1097,1142,1261,1292,1351,1472,1694,1751,1762,1844 'stop':227,1816 'structur':656 'style':590,597,602 'subag':332,1104,1301 'subsequ':931 'suggest':1128 'summari':1180,1828 'support':253,256,364,1108,1305 'tabl':585 'tag':599,604 'tailwind':586 'task/agent':333 'tc':1018 'tc-auth':1017 'tdd':53,424,969 'templat':655 'termin':299,304,819,1078,1153,1365,1858 'test':34,62,74,135,422,474,887,944,967,998,1005,1014,1030,1035,1047,1055,1061,1069,1085,1196,1198,1209,1213,1234,1239,1245,1386,1391,1423,1426,1765,1776,1783,1790,1805 'test-driven':61 'text':644 'thread':46,1549,1561,1721,1725,1910 'titl':1173 'tool':282,295,301,314,325,334,348 'topic':581 'topic-agent-skills' 'topic-ai-agents' 'topic-ai-code-review' 'topic-autonomous-coding' 'topic-autonomous-dev-team' 'topic-ci-cd' 'topic-claude-code' 'topic-claude-code-hooks' 'topic-code-review-automation' 'topic-codex-cli' 'topic-coding-agents' 'topic-devops-automation' 'toplevel':1871 'touch':961 'track':1924 'tree':674 'trigger':47,191,1491,1573,1610,1632,1727 'two':77 'type':828,1160,1174 'u':1165 'ui':31,515,615,668 'unattend':90 'unit':1029,1054,1208,1238,1385,1425 'unnecessari':1123 'unsupport':343 'updat':707,1059,1247,1251,1255,1262,1430,1807 'use':8,145,287,330,864,1102,1299,1550,1603,1607,1624,1637,1672 'user':11,112,155,176,184,523,701,985,1621,1644,1822,1836 'user-fac':522 'usernam':1494 'valid':627,1531 'valu':891 'vari':1515 'various':1464 'verif':71,229,432,1074 'verifi':634,880,910,1093 'view':1269 'viewer':316 'visual':531,622,629 'wait':130,167,452,1353,1680,1736 'want':12 'warn':925 'watch':1367,1377,1402 'whether':113 'windsurf':271 'wirefram':670 'within':952 'without':381,808 'work':75,246,782 'workflow':55,59,125,152,199,357,393,397,540 'workspac':745,792,812,958,1863 'worktre':28,66,417,482,727,738,756,769,807,813,838,845,847,856,858,892,894,901,909,911,921,926,936,940,954,1051,1847,1874,1876,1878,1881,1884 'worktrees/feat/user-authentication':903 'wrapper':197,1628 'write':33,421,966,971,1043,1052 'write/edit':324 'written':1240 'x':1283 'yarn':875 'yyyi':660 'yyyy-mm-dd':659","prices":[{"id":"ee8ca108-c25b-42ae-ac37-6da10ee44950","listingId":"0057640d-e9f5-4096-880b-5b329da01a9c","amountUsd":"0","unit":"free","nativeCurrency":null,"nativeAmount":null,"chain":null,"payTo":null,"paymentMethod":"skill-free","isPrimary":true,"details":{"org":"zxkane","category":"autonomous-dev-team","install_from":"skills.sh"},"createdAt":"2026-04-19T00:40:45.072Z"}],"sources":[{"listingId":"0057640d-e9f5-4096-880b-5b329da01a9c","source":"github","sourceId":"zxkane/autonomous-dev-team/autonomous-dev","sourceUrl":"https://github.com/zxkane/autonomous-dev-team/tree/main/skills/autonomous-dev","isPrimary":false,"firstSeenAt":"2026-04-19T00:40:45.072Z","lastSeenAt":"2026-04-22T13:03:20.329Z"}],"details":{"listingId":"0057640d-e9f5-4096-880b-5b329da01a9c","quickStartSnippet":null,"exampleRequest":null,"exampleResponse":null,"schema":null,"openapiUrl":null,"agentsTxtUrl":null,"citations":[],"useCases":[],"bestFor":[],"notFor":[],"kindDetails":{"org":"zxkane","slug":"autonomous-dev","github":{"repo":"zxkane/autonomous-dev-team","stars":15,"topics":["agent-skills","ai-agents","ai-code-review","autonomous-coding","autonomous-dev-team","ci-cd","claude-code","claude-code-hooks","code-review-automation","codex-cli","coding-agents","devops-automation","github-automation","github-issues","kiro-cli","llm-agents","openclaw","pull-request-automation","tdd","template-project"],"license":null,"html_url":"https://github.com/zxkane/autonomous-dev-team","pushed_at":"2026-04-20T15:45:13Z","description":"Turns GitHub issues into merged PRs with zero human intervention. Powered by OpenClaw, supports Claude Code, Codex CLI, and Kiro CLI.","skill_md_sha":"32b38c6d5284ef6ae520cfc81631f2072c251a82","skill_md_path":"skills/autonomous-dev/SKILL.md","default_branch":"main","skill_tree_url":"https://github.com/zxkane/autonomous-dev-team/tree/main/skills/autonomous-dev"},"layout":"multi","source":"github","category":"autonomous-dev-team","frontmatter":{"name":"autonomous-dev","description":"This skill should be used when the user wants to develop a feature, fix a bug, create a pull request, set up a git worktree, design a UI component, write test cases, push changes, check CI status, address review comments, resolve review threads, trigger bot reviews (/q review, /codex review), or follow a TDD development workflow. Covers the complete lifecycle from design canvas through worktree creation, test-first development, code review, PR creation, CI verification, reviewer bot interaction, E2E testing, and worktree cleanup. Supports interactive and fully autonomous modes for GitHub issue implementation."},"skills_sh_url":"https://skills.sh/zxkane/autonomous-dev-team/autonomous-dev"},"updatedAt":"2026-04-22T13:03:20.329Z"}}