[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"skill:01M6CW418GHEFPACNZRG7V01FQ":3,"skill-same:01M6CW418GHEFPACNZRG7V01FQ":64},{"skillNo":4,"name":5,"nameZh":6,"description":7,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":6,"repositoryNo":12,"starsCount":13,"licenseSpdxId":14,"licenseFileUrl":6,"sourceHtmlUrl":15,"categories":16,"keywords":22,"publishedAt":38,"linkOnly":39,"markdownContent":40,"markdownBody":41,"frontMatter":42,"refName":43,"commitSha":44,"skillDirectory":45,"forksCount":46,"repositoryHtmlUrl":47,"repositoryHomepageUrl":6,"ownerHtmlUrl":48,"ownerBlogUrl":6,"ownerDescription":6,"updatedAt":49,"install":50,"resources":63},"01M6CW418GHEFPACNZRG7V01FQ","debug-bug",null,"Bug fix lifecycle — investigate, classify severity, fix, test, ship. Severity determines process rigor (minor=lightweight, major\u002Fcritical=board review). Run \u002Fdebug-bug to start.","stefans71\u002FFrontierBoard","stefans71","https:\u002F\u002Favatars.githubusercontent.com\u002Fu\u002F66725311?v=4","01M6CW3VPYK13M9K13NZZBXZ36","01M6CW3VTXG8DT51A3FN2SQM1V",6,"MIT","https:\u002F\u002Fgithub.com\u002Fstefans71\u002FFrontierBoard\u002Fblob\u002Fmain\u002F.claude\u002Fskills\u002Fdebug-bug\u002FSKILL.md",[17],{"categoryNo":18,"categoryName":19,"taxonomyType":20,"taxonomyTypeDesc":21},"01JBSEEDCAT000000000000030","开发工程师","PROFESSION","适用职业",[23,24,25,26,27,28,29,30,31,32,33,34,35,36,37],"debug","bug","debug bug","review","test","run","start","severity","lifecycle","ship","investigate","bug fix","board","fix","process","2026-11-10 10:12:30",false,"---\nname: debug-bug\ndescription: Bug fix lifecycle — investigate, classify severity, fix, test, ship. Severity determines process rigor (minor=lightweight, major\u002Fcritical=board review). Run \u002Fdebug-bug to start.\n---\n\n# \u002Fdebug-bug — Bug Fix Lifecycle\n\nRun `\u002Fdebug-bug` to investigate and fix a bug with proper quality gates. The process adapts based on bug severity — minor bugs get a lightweight path, critical bugs get full board review.\n\n## Phase 0: Preflight\n\n### Read Architecture Context First\n\n**CRITICAL:** Before touching any code, read these files to understand the system:\n\n1. `CLAUDE.md` — routing, principles, review process\n2. `docs\u002FREVIEW-SOP.md` — 4-round SOP, severity levels\n3. `.board\u002Fboard\u002FBOARD.md` — agent invocations, auth strategy, isolation mode\n4. `.claude\u002Fskills\u002Fsetup\u002FSKILL.md` — Hard-Won Knowledge section (operational landmines)\n\n**Key paths (commonly confused):**\n\n- `.board\u002Fboard\u002FBOARD.md` — operational source of truth (invocations, auth, proxy)\n- `.board\u002Fboard\u002F{agent}\u002F` — agent dirs (inbox, outbox, contexts, settings)\n- `.board\u002Fboard\u002FDEFERRED_WORK.md` — active deferred items (legacy, also check tasks.json status=deferred)\n- `tasks.json` — master task index (bugs, features, deferred work)\n- `docs\u002Ftasks\u002F` — per-task YAML files with phase gates and checklists\n- `.claude\u002Fskills\u002F` — skill definitions (what the orchestrator follows)\n\n### Run Diagnostic First\n\n```bash\n# Run \u002Fdebug diagnostic script (Section 10)\n# This catches 90% of common issues before investigation starts\n```\n\nRead `.claude\u002Fskills\u002Fdebug\u002FSKILL.md` Section 10 and run the quick diagnostic script.\n\n### Clean Working Tree\n\n```bash\ngit status --porcelain\n```\n\nIf dirty, commit or stash first.\n\n### Check for In-Progress Tasks\n\n```bash\n# Check tasks.json for in-progress bugs\ncat tasks.json | grep '\"status\": \"in-progress\"'\n# Check for existing task YAML files\nls docs\u002Ftasks\u002FFB-*.yaml 2>\u002Fdev\u002Fnull\n```\n\nIf in-progress bugs exist, show them and ask: continue existing, or start new?\n\n### Create Task Entry\n\nAsk user for bug name. Assign the next available `FB-XXX` ID from `tasks.json`.\n\n1. Add an entry to `tasks.json` with `status: open`, `type: bug`\n2. Create `docs\u002Ftasks\u002F{task-id}-{name}.yaml` from `docs\u002Ftasks\u002FTEMPLATE.yaml`:\n\n```yaml\nbug: { name }\ntype: bug\nseverity: null # set in Phase 2\nreported: { today }\nstatus: pending\n\ngates:\n  investigation:\n    status: pending\n    root_cause: null\n    files_affected: []\n    touches_security: false\n    date: null\n\n  classification:\n    status: pending\n    severity: null\n    escalated: false\n    board_required: false\n    date: null\n\n  board_review:\n    status: pending\n    rounds: 0\n    findings: []\n    date: null\n\n  fix:\n    status: pending\n    files_modified: []\n    regression_test: false\n    date: null\n\n  smoke_test:\n    status: pending\n    method: null\n    result: null\n    all_agents_pass: false\n    skipped_reason: null\n    date: null\n\n  code_review:\n    status: pending\n    rounds: 0\n    findings: []\n    date: null\n\n  cleanup:\n    status: pending\n    artifacts_removed: []\n    date: null\n\n  final_signoff:\n    status: pending\n    date: null\n```\n\n## Phase 1: Investigate\n\n### Run \u002Fdebug Diagnostic\n\nStart with the diagnostic script from `\u002Fdebug` skill. This checks: BOARD.md exists, auth tokens, agent directories, lockfile, deferred items.\n\n### Reproduce the Bug\n\n1. Read the bug report \u002F user description\n2. Run a single-agent smoke test from `\u002Fdebug` skill\n3. Trace the execution path: BOARD.md invocation → `sudo -u $BOARD_USER` → CLI → agent reads CLAUDE.md → reads inbox → writes outbox\n\n### Read Relevant Source Files\n\nDon't guess — read the actual code. Follow the flow:\n- **Bare mode:** BOARD.md command → `sudo -u $BOARD_USER bash -c 'unset CLAUDECODE && ...'` → CLI reads CLAUDE.md → reads inbox → writes outbox\n- **Auth flow:** CLI reads credentials from board user's home directory → authenticates with upstream API\n\n### Document Root Cause\n\nUpdate task YAML:\n\n```yaml\ninvestigation:\n  status: done\n  root_cause: \"description of what's wrong and why\"\n  files_affected: [list of files]\n  touches_security: true\u002Ffalse # auth, permissions, blind review enforcement\n```\n\n## Phase 2: Classify Severity\n\nBased on investigation, classify:\n\n| Severity     | Criteria                                                                    | Board Review |\n| ------------ | --------------------------------------------------------------------------- | ------------ |\n| **Minor**    | UX issue, formatting, missing log output, cosmetic skill error              | Skip         |\n| **Major**    | Auth failure, agent can't produce reports, wrong model                      | Recommended  |\n| **Critical** | Credential leakage, cross-agent data access, review SOP bypass              | **Required** |\n\n### Escalation Check\n\nIf the fix touches ANY of these, escalate severity by one level:\n\n- Agent invocation pattern (BOARD.md commands, parallelism, timeout)\n- Review SOP (4-round process, lockfile, deferred items, blind review)\n- Setup skill templates (invocation commands that get written to BOARD.md)\n\nUpdate task YAML: `classification.status: done`, `severity: minor|major|critical`\n\n## Phase 3: Board Review (Major\u002FCritical only)\n\nFor **minor** bugs: set `board_review.status: skipped` and proceed to Phase 4.\n\nFor **major\u002Fcritical** bugs: submit investigation + proposed fix to the board.\n\n**Brief must include:**\n\n- Bug description and reproduction steps\n- Root cause analysis\n- Proposed fix with code snippets\n- Files affected\n- Why the bug matters (auth impact, agent failure, data risk)\n\n**Round structure:** Same as review SOP (blind → consolidation → deliberation → confirmation).\n\n**Retry enforcement:** If any agent fails to produce a report in any round, retry that agent at least once before proceeding. If the retry also fails, do NOT silently continue — inform the user and get explicit approval to proceed with reduced coverage. With 3 agents, losing one means the tiebreaker is gone. Log all failures and retries in the task YAML.\n\nUpdate task YAML: `board_review.status: done`, `rounds: N`\n\n## Phase 4: Fix\n\nImplement the fix. For each change, document what file was modified and why.\n\n**Verification during fix (not after):**\n- Run `\u002Fdebug` diagnostic after each significant change\n- Single-agent smoke test after auth changes\n- Check BOARD.md invocation commands match actual bare-mode commands being tested\n\nUpdate task YAML: `fix.status: done`, `files_modified: [...]`\n\n## Phase 5: Smoke Test\n\n**DO NOT skip this. ALL agents must pass.**\n\n### Full Smoke Test (Bare Mode)\n\n```bash\nBOARD=\u002Fpath\u002Fto\u002FFrontierBoard\u002F.board\nBOARD_USER=llmuser  # or whatever board user was created during setup\n\n# Write test brief to all agents\nfor agent in pragmatist systems-thinker skeptic; do\n  echo \"Smoke test: Confirm identity. Write one sentence to outbox\u002Freport.md.\" > $BOARD\u002Fboard\u002F$agent\u002Finbox\u002Fbrief.md\n  echo \"Smoke test context.\" > $BOARD\u002Fboard\u002F$agent\u002Finbox\u002Fcontext.md\n  rm -f $BOARD\u002Fboard\u002F$agent\u002Foutbox\u002Freport.md\ndone\n\n# Run each agent (copy parallelism pattern from BOARD.md)\nfor agent in pragmatist systems-thinker skeptic; do\n  AGENT_DIR=\"$BOARD\u002Fboard\u002F$agent\"\n  sudo -u $BOARD_USER bash -c \"unset CLAUDECODE && cd $AGENT_DIR && claude --dangerously-skip-permissions -p 'Read inbox\u002Fbrief.md. Follow its instructions. Write your response to outbox\u002Freport.md.'\" &\ndone\nwait\n\n# Verify ALL reports\nfor agent in pragmatist systems-thinker skeptic; do\n  [ -s \"$BOARD\u002Fboard\u002F$agent\u002Foutbox\u002Freport.md\" ] && echo \"PASS: $agent\" || echo \"FAIL: $agent\"\ndone\n```\n\n**Every agent must produce a report. If any agent fails, the bug is not fixed.**\n\nUpdate task YAML:\n\n```yaml\nsmoke_test:\n  status: done\n  method: 'all 3 agents in parallel, bare mode'\n  result: 'all passed \u002F agent X failed with ...'\n  all_agents_pass: true\u002Ffalse\n```\n\n## Phase 6: Code Review (Major\u002FCritical only)\n\nFor **minor** bugs: set `code_review.status: skipped` and proceed to Phase 7.\n\nFor **major\u002Fcritical** bugs: send `git diff` to the board.\n\nUpdate task YAML: `code_review.status: done`, `rounds: N`\n\n## Phase 7: Ship\n\n1. Commit with bug reference: `git add -A && git commit -m \"fix: {description}\"`\n2. Update task YAML: `final_signoff.status: approved`, `status: approved`\n\n## Phase 8: Cleanup\n\nRemove temporary artifacts created during debugging. **The bug checklist stays** (it's the audit trail).\n\n### Artifacts to Remove\n\n```bash\nBOARD=\u002Fpath\u002Fto\u002FFrontierBoard\u002F.board\n\n# Smoke test briefs and reports (not real review artifacts)\nfor agent in $BOARD\u002Fboard\u002F*\u002F; do\n  name=$(basename \"$agent\")\n  # Only clean if brief is a smoke test\n  grep -q \"Smoke test\" \"$agent\u002Finbox\u002Fbrief.md\" 2>\u002Fdev\u002Fnull && rm -f \"$agent\u002Finbox\u002Fbrief.md\" \"$agent\u002Finbox\u002Fcontext.md\" \"$agent\u002Foutbox\u002Freport.md\"\ndone\n\n# Stale review lockfile (if left from crashed test)\nrm -f $BOARD\u002F.review-lock\n```\n\nUpdate task YAML: `cleanup.status: done`, `artifacts_removed: [list]`\n\n---\n\n## Severity Examples\n\n**Minor:**\n\n- Agent report formatting inconsistent\n- Diagnostic script false positive on a directory\n- Skill documentation typo\n\n**Major:**\n\n- Agent auth fails (OAuth token race condition, expired credentials)\n- Codex config format incompatible with new CLI version\n- Agent model not supported with ChatGPT account\n\n**Critical:**\n\n- Credential leakage to unauthorized processes\n- Agent reads sibling agent's outbox (blind review violation)\n- OAuth token written to shared location accessible by all agents\n- Review lockfile bypass allows concurrent reviews corrupting reports\n","# \u002Fdebug-bug — Bug Fix Lifecycle\n\nRun `\u002Fdebug-bug` to investigate and fix a bug with proper quality gates. The process adapts based on bug severity — minor bugs get a lightweight path, critical bugs get full board review.\n\n## Phase 0: Preflight\n\n### Read Architecture Context First\n\n**CRITICAL:** Before touching any code, read these files to understand the system:\n\n1. `CLAUDE.md` — routing, principles, review process\n2. `docs\u002FREVIEW-SOP.md` — 4-round SOP, severity levels\n3. `.board\u002Fboard\u002FBOARD.md` — agent invocations, auth strategy, isolation mode\n4. `.claude\u002Fskills\u002Fsetup\u002FSKILL.md` — Hard-Won Knowledge section (operational landmines)\n\n**Key paths (commonly confused):**\n\n- `.board\u002Fboard\u002FBOARD.md` — operational source of truth (invocations, auth, proxy)\n- `.board\u002Fboard\u002F{agent}\u002F` — agent dirs (inbox, outbox, contexts, settings)\n- `.board\u002Fboard\u002FDEFERRED_WORK.md` — active deferred items (legacy, also check tasks.json status=deferred)\n- `tasks.json` — master task index (bugs, features, deferred work)\n- `docs\u002Ftasks\u002F` — per-task YAML files with phase gates and checklists\n- `.claude\u002Fskills\u002F` — skill definitions (what the orchestrator follows)\n\n### Run Diagnostic First\n\n```bash\n# Run \u002Fdebug diagnostic script (Section 10)\n# This catches 90% of common issues before investigation starts\n```\n\nRead `.claude\u002Fskills\u002Fdebug\u002FSKILL.md` Section 10 and run the quick diagnostic script.\n\n### Clean Working Tree\n\n```bash\ngit status --porcelain\n```\n\nIf dirty, commit or stash first.\n\n### Check for In-Progress Tasks\n\n```bash\n# Check tasks.json for in-progress bugs\ncat tasks.json | grep '\"status\": \"in-progress\"'\n# Check for existing task YAML files\nls docs\u002Ftasks\u002FFB-*.yaml 2>\u002Fdev\u002Fnull\n```\n\nIf in-progress bugs exist, show them and ask: continue existing, or start new?\n\n### Create Task Entry\n\nAsk user for bug name. Assign the next available `FB-XXX` ID from `tasks.json`.\n\n1. Add an entry to `tasks.json` with `status: open`, `type: bug`\n2. Create `docs\u002Ftasks\u002F{task-id}-{name}.yaml` from `docs\u002Ftasks\u002FTEMPLATE.yaml`:\n\n```yaml\nbug: { name }\ntype: bug\nseverity: null # set in Phase 2\nreported: { today }\nstatus: pending\n\ngates:\n  investigation:\n    status: pending\n    root_cause: null\n    files_affected: []\n    touches_security: false\n    date: null\n\n  classification:\n    status: pending\n    severity: null\n    escalated: false\n    board_required: false\n    date: null\n\n  board_review:\n    status: pending\n    rounds: 0\n    findings: []\n    date: null\n\n  fix:\n    status: pending\n    files_modified: []\n    regression_test: false\n    date: null\n\n  smoke_test:\n    status: pending\n    method: null\n    result: null\n    all_agents_pass: false\n    skipped_reason: null\n    date: null\n\n  code_review:\n    status: pending\n    rounds: 0\n    findings: []\n    date: null\n\n  cleanup:\n    status: pending\n    artifacts_removed: []\n    date: null\n\n  final_signoff:\n    status: pending\n    date: null\n```\n\n## Phase 1: Investigate\n\n### Run \u002Fdebug Diagnostic\n\nStart with the diagnostic script from `\u002Fdebug` skill. This checks: BOARD.md exists, auth tokens, agent directories, lockfile, deferred items.\n\n### Reproduce the Bug\n\n1. Read the bug report \u002F user description\n2. Run a single-agent smoke test from `\u002Fdebug` skill\n3. Trace the execution path: BOARD.md invocation → `sudo -u $BOARD_USER` → CLI → agent reads CLAUDE.md → reads inbox → writes outbox\n\n### Read Relevant Source Files\n\nDon't guess — read the actual code. Follow the flow:\n- **Bare mode:** BOARD.md command → `sudo -u $BOARD_USER bash -c 'unset CLAUDECODE && ...'` → CLI reads CLAUDE.md → reads inbox → writes outbox\n- **Auth flow:** CLI reads credentials from board user's home directory → authenticates with upstream API\n\n### Document Root Cause\n\nUpdate task YAML:\n\n```yaml\ninvestigation:\n  status: done\n  root_cause: \"description of what's wrong and why\"\n  files_affected: [list of files]\n  touches_security: true\u002Ffalse # auth, permissions, blind review enforcement\n```\n\n## Phase 2: Classify Severity\n\nBased on investigation, classify:\n\n| Severity     | Criteria                                                                    | Board Review |\n| ------------ | --------------------------------------------------------------------------- | ------------ |\n| **Minor**    | UX issue, formatting, missing log output, cosmetic skill error              | Skip         |\n| **Major**    | Auth failure, agent can't produce reports, wrong model                      | Recommended  |\n| **Critical** | Credential leakage, cross-agent data access, review SOP bypass              | **Required** |\n\n### Escalation Check\n\nIf the fix touches ANY of these, escalate severity by one level:\n\n- Agent invocation pattern (BOARD.md commands, parallelism, timeout)\n- Review SOP (4-round process, lockfile, deferred items, blind review)\n- Setup skill templates (invocation commands that get written to BOARD.md)\n\nUpdate task YAML: `classification.status: done`, `severity: minor|major|critical`\n\n## Phase 3: Board Review (Major\u002FCritical only)\n\nFor **minor** bugs: set `board_review.status: skipped` and proceed to Phase 4.\n\nFor **major\u002Fcritical** bugs: submit investigation + proposed fix to the board.\n\n**Brief must include:**\n\n- Bug description and reproduction steps\n- Root cause analysis\n- Proposed fix with code snippets\n- Files affected\n- Why the bug matters (auth impact, agent failure, data risk)\n\n**Round structure:** Same as review SOP (blind → consolidation → deliberation → confirmation).\n\n**Retry enforcement:** If any agent fails to produce a report in any round, retry that agent at least once before proceeding. If the retry also fails, do NOT silently continue — inform the user and get explicit approval to proceed with reduced coverage. With 3 agents, losing one means the tiebreaker is gone. Log all failures and retries in the task YAML.\n\nUpdate task YAML: `board_review.status: done`, `rounds: N`\n\n## Phase 4: Fix\n\nImplement the fix. For each change, document what file was modified and why.\n\n**Verification during fix (not after):**\n- Run `\u002Fdebug` diagnostic after each significant change\n- Single-agent smoke test after auth changes\n- Check BOARD.md invocation commands match actual bare-mode commands being tested\n\nUpdate task YAML: `fix.status: done`, `files_modified: [...]`\n\n## Phase 5: Smoke Test\n\n**DO NOT skip this. ALL agents must pass.**\n\n### Full Smoke Test (Bare Mode)\n\n```bash\nBOARD=\u002Fpath\u002Fto\u002FFrontierBoard\u002F.board\nBOARD_USER=llmuser  # or whatever board user was created during setup\n\n# Write test brief to all agents\nfor agent in pragmatist systems-thinker skeptic; do\n  echo \"Smoke test: Confirm identity. Write one sentence to outbox\u002Freport.md.\" > $BOARD\u002Fboard\u002F$agent\u002Finbox\u002Fbrief.md\n  echo \"Smoke test context.\" > $BOARD\u002Fboard\u002F$agent\u002Finbox\u002Fcontext.md\n  rm -f $BOARD\u002Fboard\u002F$agent\u002Foutbox\u002Freport.md\ndone\n\n# Run each agent (copy parallelism pattern from BOARD.md)\nfor agent in pragmatist systems-thinker skeptic; do\n  AGENT_DIR=\"$BOARD\u002Fboard\u002F$agent\"\n  sudo -u $BOARD_USER bash -c \"unset CLAUDECODE && cd $AGENT_DIR && claude --dangerously-skip-permissions -p 'Read inbox\u002Fbrief.md. Follow its instructions. Write your response to outbox\u002Freport.md.'\" &\ndone\nwait\n\n# Verify ALL reports\nfor agent in pragmatist systems-thinker skeptic; do\n  [ -s \"$BOARD\u002Fboard\u002F$agent\u002Foutbox\u002Freport.md\" ] && echo \"PASS: $agent\" || echo \"FAIL: $agent\"\ndone\n```\n\n**Every agent must produce a report. If any agent fails, the bug is not fixed.**\n\nUpdate task YAML:\n\n```yaml\nsmoke_test:\n  status: done\n  method: 'all 3 agents in parallel, bare mode'\n  result: 'all passed \u002F agent X failed with ...'\n  all_agents_pass: true\u002Ffalse\n```\n\n## Phase 6: Code Review (Major\u002FCritical only)\n\nFor **minor** bugs: set `code_review.status: skipped` and proceed to Phase 7.\n\nFor **major\u002Fcritical** bugs: send `git diff` to the board.\n\nUpdate task YAML: `code_review.status: done`, `rounds: N`\n\n## Phase 7: Ship\n\n1. Commit with bug reference: `git add -A && git commit -m \"fix: {description}\"`\n2. Update task YAML: `final_signoff.status: approved`, `status: approved`\n\n## Phase 8: Cleanup\n\nRemove temporary artifacts created during debugging. **The bug checklist stays** (it's the audit trail).\n\n### Artifacts to Remove\n\n```bash\nBOARD=\u002Fpath\u002Fto\u002FFrontierBoard\u002F.board\n\n# Smoke test briefs and reports (not real review artifacts)\nfor agent in $BOARD\u002Fboard\u002F*\u002F; do\n  name=$(basename \"$agent\")\n  # Only clean if brief is a smoke test\n  grep -q \"Smoke test\" \"$agent\u002Finbox\u002Fbrief.md\" 2>\u002Fdev\u002Fnull && rm -f \"$agent\u002Finbox\u002Fbrief.md\" \"$agent\u002Finbox\u002Fcontext.md\" \"$agent\u002Foutbox\u002Freport.md\"\ndone\n\n# Stale review lockfile (if left from crashed test)\nrm -f $BOARD\u002F.review-lock\n```\n\nUpdate task YAML: `cleanup.status: done`, `artifacts_removed: [list]`\n\n---\n\n## Severity Examples\n\n**Minor:**\n\n- Agent report formatting inconsistent\n- Diagnostic script false positive on a directory\n- Skill documentation typo\n\n**Major:**\n\n- Agent auth fails (OAuth token race condition, expired credentials)\n- Codex config format incompatible with new CLI version\n- Agent model not supported with ChatGPT account\n\n**Critical:**\n\n- Credential leakage to unauthorized processes\n- Agent reads sibling agent's outbox (blind review violation)\n- OAuth token written to shared location accessible by all agents\n- Review lockfile bypass allows concurrent reviews corrupting reports\n",{"name":5,"description":7},"main","e01303ad28862b5c26c950358250aeb4f74090e6",".claude\u002Fskills\u002Fdebug-bug",1,"https:\u002F\u002Fgithub.com\u002Fstefans71\u002FFrontierBoard","https:\u002F\u002Fgithub.com\u002Fstefans71","2026-04-20 07:23:08",{"source":8,"skillName":5,"commands":51,"prompt":61,"zipAvailable":62},[52,55,58],{"tool":53,"command":54},"npx","npx skills add stefans71\u002FFrontierBoard --skill debug-bug",{"tool":56,"command":57},"bunx","bunx skills add stefans71\u002FFrontierBoard --skill debug-bug",{"tool":59,"command":60},"pnpm","pnpm dlx skills add stefans71\u002FFrontierBoard --skill debug-bug","安装技能，技能地址：https:\u002F\u002Fgithub.com\u002Fstefans71\u002FFrontierBoard\u002Fblob\u002Fmain\u002F.claude\u002Fskills\u002Fdebug-bug\u002FSKILL.md",true,[],[65,78,95,107,117,134],{"skillNo":66,"name":67,"nameZh":6,"description":68,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":6,"repositoryNo":12,"starsCount":13,"licenseSpdxId":14,"licenseFileUrl":6,"sourceHtmlUrl":69,"categories":70,"keywords":71,"publishedAt":38},"01M6CW5BKYDY1PC8YPCHNZN1V5","teardown","Remove a FrontierBoard installation. Run when the user types \u002Fteardown or asks to uninstall, remove, or clean up the board.","https:\u002F\u002Fgithub.com\u002Fstefans71\u002FFrontierBoard\u002Fblob\u002Fmain\u002F.claude\u002Fskills\u002Fteardown\u002FSKILL.md",[],[67,72,28,35,73,74,75,76,77],"uninstall","user","remove","up","clean","installation",{"skillNo":79,"name":80,"nameZh":6,"description":81,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":6,"repositoryNo":12,"starsCount":13,"licenseSpdxId":14,"licenseFileUrl":6,"sourceHtmlUrl":82,"categories":83,"keywords":89,"publishedAt":38},"01M6CW599G8TXF87JTYRPHRKJ7","smoke-test","Validate a board installation works end-to-end. Run when the user types \u002Fsmoke-test or asks to test, validate, or verify the board setup.","https:\u002F\u002Fgithub.com\u002Fstefans71\u002FFrontierBoard\u002Fblob\u002Fmain\u002F.claude\u002Fskills\u002Fsmoke-test\u002FSKILL.md",[84],{"categoryNo":85,"categoryName":86,"taxonomyType":87,"taxonomyTypeDesc":88},"01JBSEEDCAT000000000000008","测试与质量","SCENARIO","使用场景",[90,91,27,73,92,93,28,94,35,77],"smoke test","smoke","validate","setup","verify",{"skillNo":96,"name":93,"nameZh":6,"description":97,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":6,"repositoryNo":12,"starsCount":13,"licenseSpdxId":14,"licenseFileUrl":6,"sourceHtmlUrl":98,"categories":99,"keywords":100,"publishedAt":38},"01M6CW56Q6B52QVY7YFDXJCDP2","Set up the Board of Governance. Run when the user types \u002Fsetup or asks to set up, install, or configure the board for the first time.","https:\u002F\u002Fgithub.com\u002Fstefans71\u002FFrontierBoard\u002Fblob\u002Fmain\u002F.claude\u002Fskills\u002Fsetup\u002FSKILL.md",[],[93,35,73,101,75,102,103,104,105,28,106],"install","time","first","configure","governance","SET",{"skillNo":108,"name":28,"nameZh":6,"description":109,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":6,"repositoryNo":12,"starsCount":13,"licenseSpdxId":14,"licenseFileUrl":6,"sourceHtmlUrl":110,"categories":111,"keywords":112,"publishedAt":38},"01M6CW5435Q93Y2Q1427PSDWMA","Run the board agents and synthesise their reports. Run when the user types \u002Frun, asks to run the board, or after \u002Fbrief when the user confirms they're ready to go.","https:\u002F\u002Fgithub.com\u002Fstefans71\u002FFrontierBoard\u002Fblob\u002Fmain\u002F.claude\u002Fskills\u002Frun\u002FSKILL.md",[],[28,113,114,35,73,115,116],"go","brief","ready","reports",{"skillNo":118,"name":119,"nameZh":6,"description":120,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":6,"repositoryNo":12,"starsCount":13,"licenseSpdxId":14,"licenseFileUrl":6,"sourceHtmlUrl":121,"categories":122,"keywords":123,"publishedAt":38},"01M6CW518M6QPGGW3DAT6JNEZQ","review-release","Review a GitHub repo or release using the board. Use when the user types \u002Freview-release, asks to \"review a release\", \"check if this is safe to install\", \"look at the latest version of [repo]\", wants to \"try installing it and see what breaks\", or any variant of reviewing an external GitHub project before contributing to it or installing it.","https:\u002F\u002Fgithub.com\u002Fstefans71\u002FFrontierBoard\u002Fblob\u002Fmain\u002F.claude\u002Fskills\u002Freview-release\u002FSKILL.md",[],[124,125,26,126,127,128,129,130,131,35,73,101,132,133],"review release","release","look","safe","github","contributing","repo","check","use","version",{"skillNo":135,"name":136,"nameZh":6,"description":137,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":6,"repositoryNo":12,"starsCount":13,"licenseSpdxId":14,"licenseFileUrl":6,"sourceHtmlUrl":138,"categories":139,"keywords":143,"publishedAt":38},"01M6CW4YSSPY6FXCR6DMZWF9XX","project-tests","Generate test specs (Skeptic writes) and verify test code (Skeptic reviews). Two modes — \u002Fproject tests for specs, \u002Fproject tests --verify for code review.","https:\u002F\u002Fgithub.com\u002Fstefans71\u002FFrontierBoard\u002Fblob\u002Fmain\u002F.claude\u002Fskills\u002Fproject\u002Ftests\u002FSKILL.md",[140],{"categoryNo":141,"categoryName":142,"taxonomyType":87,"taxonomyTypeDesc":88},"01JBSEEDCAT000000000000007","代码审查",[144,145,146,27,26,147,148,94,149],"project tests","code","generate","verify test","skeptic","code review"]