[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"skill:01M7F4X23YWRH2KXJ6TYVTYDZ3":3,"skill-same:01M7F4X23YWRH2KXJ6TYVTYDZ3":56},{"skillNo":4,"name":5,"nameZh":6,"description":7,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":16,"categories":17,"keywords":18,"publishedAt":19,"linkOnly":20,"markdownContent":21,"markdownBody":22,"frontMatter":23,"refName":35,"commitSha":36,"skillDirectory":37,"forksCount":38,"repositoryHtmlUrl":39,"repositoryHomepageUrl":6,"ownerHtmlUrl":40,"ownerBlogUrl":6,"ownerDescription":41,"updatedAt":42,"install":43,"resources":55},"01M7F4X23YWRH2KXJ6TYVTYDZ3","skill-factory",null,"Run a full build-and-ship pipeline from a spec — use for hands-off project generation","lza6\u002FArmor-piercing-system-prompts","lza6","https:\u002F\u002Favatars.githubusercontent.com\u002Fu\u002F128880206?v=4","01M7EB3K7WPX5T28EJ2NVHE6EG","Chinese-tingfeng","01M7EB3K97E1KQ16QWKJMFNGX8",11,"MIT","https:\u002F\u002Fgithub.com\u002Flza6\u002FArmor-piercing-system-prompts\u002Fblob\u002Fmain\u002F05-技能与代理开发\u002Fskill-factory-claude-octopus\u002FSKILL.md",[],[],"2026-11-16 10:05:34",false,"---\nname: skill-factory\ndisable-model-invocation: true\naliases:\n  - factory\n  - dark-factory\n  - build-from-spec\n  - autonomous-build\ndescription: \"Run a full build-and-ship pipeline from a spec — use for hands-off project generation\"\nexecution_mode: enforced\nvalidation_gates:\n  - spec_file_validated\n  - orchestrate_sh_executed\n  - factory_report_exists\n---\n\n# STOP - SKILL ALREADY LOADED\nDO NOT call Skill() again. Execute directly.\n\n## EXECUTION CONTRACT (MANDATORY — 8-step sequence, CANNOT SKIP)\n\n### STEP 1: Clarifying Questions (MANDATORY)\nAsk via AskUserQuestion BEFORE any other action:\n1. Spec location — path to NLSpec file, or paste inline\n2. Satisfaction target override (Use spec default \u002F Custom 0.80-0.99)\n3. Cost approval — ~$0.50-2.00 for ~20-30 agent calls (Approve \u002F Approve --ci \u002F Decline)\n\nIf spec path provided inline with the command, use it but still ask remaining questions.\nIf user says \"skip\", use defaults and proceed.\nDO NOT proceed to Step 2 until answered.\n\n### STEP 2: Display Visual Indicators (MANDATORY — BLOCKING)\nCheck providers:\n```bash\ncommand -v codex &> \u002Fdev\u002Fnull && codex_status=\"Available\" || codex_status=\"Not installed\"\ncommand -v agy &> \u002Fdev\u002Fnull && agy_status=\"Available\" || agy_status=\"Not installed\"\n```\nDisplay banner:\n```\n🐙 CLAUDE OCTOPUS ACTIVATED - Dark Factory Mode\nPipeline: Parse → Scenarios → Embrace → Holdout → Score → Report\n\nProviders:\n  Codex CLI - \u003Cstatus> — Scenario generation + holdout evaluation\n  Antigravity CLI - \u003Cstatus> — Additional external-model challenge\n  Claude - Orchestration + satisfaction scoring\n\nSpec: \u003Cpath>\nEstimated: $0.50-2.00 \u002F 15-45 min\n```\n\nValidation: All external providers unavailable → continue with Claude-only (warn user about reduced diversity). At least one available → proceed normally.\n\n### STEP 3: Validate Spec File (MANDATORY — Validation Gate)\n```bash\nif [[ ! -f \"\u003Cspec_path>\" ]]; then\n    echo \"ERROR: Spec file not found at \u003Cspec_path>\"\n    exit 1\nfi\n# Check minimum content\nword_count=$(wc -w \u003C \"\u003Cspec_path>\")\nif [[ $word_count -lt 20 ]]; then\n    echo \"WARNING: Spec is very thin ($word_count words). Results may be limited.\"\nfi\n```\nIf spec file missing → STOP, ask user for correct path.\nIf spec is thin (\u003C 50 words) → WARN but proceed.\nDO NOT proceed past this gate if file does not exist.\n\n### STEP 4: Read Prior State (OPTIONAL)\n```bash\n\"${HOME}\u002F.claude-octopus\u002Fplugin\u002Fscripts\u002Fstate-manager.sh\" init_state 2>\u002Fdev\u002Fnull || true\n\"${HOME}\u002F.claude-octopus\u002Fplugin\u002Fscripts\u002Fstate-manager.sh\" set_current_workflow \"factory\" \"factory\" 2>\u002Fdev\u002Fnull || true\n```\nFailure → continue without state, warn user.\n\n### STEP 4.5: Adversarial Scenario Coverage Gate (RECOMMENDED)\n\n**Before committing to the expensive embrace phase (~$0.50-2.00), verify that generated scenarios actually cover the spec's edge cases.** A quick cross-provider challenge here can prevent wasting an entire factory run on incomplete scenario coverage.\n\n**After orchestrate.sh parses the spec (Phase 1-2) and before embrace execution (Phase 4), dispatch a scenario coverage review:**\n\nIf a second provider is available, dispatch the challenge through Octopus routing:\n\n```bash\n# Read the spec to extract behaviors and constraints\nSPEC_CONTENT=$(\u003C\"\u003Cspec_path>\")\nreview_provider=\"\"\ncommand -v codex >\u002Fdev\u002Fnull 2>&1 && review_provider=\"codex\"\n[[ -z \"$review_provider\" ]] && command -v agy >\u002Fdev\u002Fnull 2>&1 && review_provider=\"agy\"\n\nif [[ -n \"$review_provider\" ]]; then\n  \"${HOME}\u002F.claude-octopus\u002Fplugin\u002Fscripts\u002Forchestrate.sh\" spawn \"$review_provider\" \\\n    \"You are a QA adversary. Given this specification and generated scenarios, identify coverage gaps.\n\nSPECIFICATION:\n${SPEC_CONTENT}\n\nAnswer:\n1. Which spec BEHAVIORS have no scenario testing them?\n2. Which EDGE CASES from the spec constraints are untested?\n3. Which FAILURE MODES are not covered (network failures, invalid input, concurrent access)?\n4. Rate overall coverage: SUFFICIENT \u002F GAPS-FOUND \u002F CRITICAL-GAPS\n\nBe specific — cite the behavior or constraint ID that lacks coverage.\" 2>\u002Fdev\u002Fnull || true\nfi\n```\n\n**After receiving the challenge response:**\n- If CRITICAL-GAPS found: warn the user and suggest refining the spec with `\u002Focto:spec` before proceeding\n- If GAPS-FOUND: note the gaps but proceed — the holdout phase (Phase 5) will catch some of these\n- If SUFFICIENT: proceed with confidence\n\n**This is a lightweight gate — it adds ~30 seconds but can save a $2.00 factory run on a spec with poor scenario coverage.**\n\n**Skip with `--fast` or when user explicitly requests speed over thoroughness.**\n\n---\n\n### STEP 5: Execute orchestrate.sh factory (MANDATORY — Bash Tool)\n```bash\n${HOME}\u002F.claude-octopus\u002Fplugin\u002Fscripts\u002Forchestrate.sh factory --spec \"\u003Cspec_path>\"\n```\n\nWith optional flags based on Step 1 answers:\n- `--holdout-ratio \u003Cvalue>` if custom ratio requested\n- `--max-retries \u003Cvalue>` if custom retry count\n- `--ci` if user approved non-interactive mode\n\n\u003CHARD-GATE>\nPROHIBITED from:\n- Running embrace directly (MUST use factory command which wraps it)\n- Simulating or faking holdout testing\n- Substituting direct Claude analysis for multi-provider scoring\n- Skipping the factory pipeline\n- Creating working\u002Fprogress files in plugin directory\n\u003C\u002FHARD-GATE>\n\n### STEP 6: Verify Factory Report (MANDATORY — Validation Gate)\n```bash\nREPORT_FILE=$(find .octo\u002Ffactory -name \"factory-report.md\" -mmin -60 2>\u002Fdev\u002Fnull | sort -r | head -1)\nif [[ -z \"$REPORT_FILE\" ]]; then\n    echo \"ERROR: Factory report not found\"\n    exit 1\nfi\ncat \"$REPORT_FILE\"\n```\nIf validation fails: report error, show logs from `~\u002F.claude-octopus\u002Flogs\u002F`, DO NOT proceed, DO NOT substitute.\n\n### STEP 7: Read Scores and Present Results (MANDATORY)\n```bash\nSCORES_FILE=$(find .octo\u002Ffactory -name \"satisfaction-scores.json\" -mmin -60 2>\u002Fdev\u002Fnull | sort -r | head -1)\nif [[ -n \"$SCORES_FILE\" ]]; then\n    cat \"$SCORES_FILE\"\nfi\n```\n\nPresent to user:\n1. **Verdict** with emoji (PASS\u002FWARN\u002FFAIL)\n2. **Composite score** vs satisfaction target\n3. **Dimension breakdown** (behavior, constraints, holdout, quality)\n4. **Holdout highlights** — which blind scenarios passed\u002Ffailed\n5. **Artifact directory** path for deep review\n\n### STEP 8: Update State & Next Steps (MANDATORY)\n```bash\n\"${HOME}\u002F.claude-octopus\u002Fplugin\u002Fscripts\u002Fstate-manager.sh\" record_decision \"factory\" \"Factory run completed: \u003Cverdict> (\u003Cscore>\u002F\u003Ctarget>)\" 2>\u002Fdev\u002Fnull || true\n```\n\nPresent next-step suggestions based on verdict:\n- **PASS:** Implementation meets spec. Review artifacts, run manual testing, ship.\n- **WARN:** Close to target. Review holdout failures, consider targeted fixes.\n- **FAIL:** Below target. Review `holdout-results.md` for specific failures. Consider:\n  - Refining the NLSpec with `\u002Focto:spec`\n  - Manual fixes + re-run with `--max-retries 2`\n  - Breaking spec into smaller, clearer pieces\n\nInclude attribution footer:\n```\nDark Factory Mode powered by Claude Octopus v8.25.0\nPipeline: Spec → Scenarios → Embrace → Holdout → Score → Report\nProviders: Codex | Antigravity | Claude\n```\n\n## Error Handling (by step)\n- Step 1 (Questions): If user declines all, proceed with defaults\n- Step 2 (Providers): Both external unavailable → Claude-only mode (warn)\n- Step 3 (Spec): File missing → STOP. Thin spec → WARN and proceed.\n- Step 4 (State): Failure → continue without state\n- Step 5 (orchestrate.sh): Show bash error, check logs — DO NOT substitute\n- Step 6 (Report): Missing → show logs, DO NOT proceed\n- Step 7 (Scores): Missing JSON → extract from report markdown\n- Step 8 (State): Failure → skip state update, still present results\n\n## Prohibited Actions\n- CANNOT skip orchestrate.sh factory execution\n- CANNOT simulate or fake the factory pipeline\n- CANNOT substitute direct Claude analysis for multi-provider scoring\n- CANNOT skip spec validation gate\n- CANNOT proceed past a failed validation gate\n- CANNOT create working\u002Fprogress files in plugin directory\n","# STOP - SKILL ALREADY LOADED\nDO NOT call Skill() again. Execute directly.\n\n## EXECUTION CONTRACT (MANDATORY — 8-step sequence, CANNOT SKIP)\n\n### STEP 1: Clarifying Questions (MANDATORY)\nAsk via AskUserQuestion BEFORE any other action:\n1. Spec location — path to NLSpec file, or paste inline\n2. Satisfaction target override (Use spec default \u002F Custom 0.80-0.99)\n3. Cost approval — ~$0.50-2.00 for ~20-30 agent calls (Approve \u002F Approve --ci \u002F Decline)\n\nIf spec path provided inline with the command, use it but still ask remaining questions.\nIf user says \"skip\", use defaults and proceed.\nDO NOT proceed to Step 2 until answered.\n\n### STEP 2: Display Visual Indicators (MANDATORY — BLOCKING)\nCheck providers:\n```bash\ncommand -v codex &> \u002Fdev\u002Fnull && codex_status=\"Available\" || codex_status=\"Not installed\"\ncommand -v agy &> \u002Fdev\u002Fnull && agy_status=\"Available\" || agy_status=\"Not installed\"\n```\nDisplay banner:\n```\n🐙 CLAUDE OCTOPUS ACTIVATED - Dark Factory Mode\nPipeline: Parse → Scenarios → Embrace → Holdout → Score → Report\n\nProviders:\n  Codex CLI - \u003Cstatus> — Scenario generation + holdout evaluation\n  Antigravity CLI - \u003Cstatus> — Additional external-model challenge\n  Claude - Orchestration + satisfaction scoring\n\nSpec: \u003Cpath>\nEstimated: $0.50-2.00 \u002F 15-45 min\n```\n\nValidation: All external providers unavailable → continue with Claude-only (warn user about reduced diversity). At least one available → proceed normally.\n\n### STEP 3: Validate Spec File (MANDATORY — Validation Gate)\n```bash\nif [[ ! -f \"\u003Cspec_path>\" ]]; then\n    echo \"ERROR: Spec file not found at \u003Cspec_path>\"\n    exit 1\nfi\n# Check minimum content\nword_count=$(wc -w \u003C \"\u003Cspec_path>\")\nif [[ $word_count -lt 20 ]]; then\n    echo \"WARNING: Spec is very thin ($word_count words). Results may be limited.\"\nfi\n```\nIf spec file missing → STOP, ask user for correct path.\nIf spec is thin (\u003C 50 words) → WARN but proceed.\nDO NOT proceed past this gate if file does not exist.\n\n### STEP 4: Read Prior State (OPTIONAL)\n```bash\n\"${HOME}\u002F.claude-octopus\u002Fplugin\u002Fscripts\u002Fstate-manager.sh\" init_state 2>\u002Fdev\u002Fnull || true\n\"${HOME}\u002F.claude-octopus\u002Fplugin\u002Fscripts\u002Fstate-manager.sh\" set_current_workflow \"factory\" \"factory\" 2>\u002Fdev\u002Fnull || true\n```\nFailure → continue without state, warn user.\n\n### STEP 4.5: Adversarial Scenario Coverage Gate (RECOMMENDED)\n\n**Before committing to the expensive embrace phase (~$0.50-2.00), verify that generated scenarios actually cover the spec's edge cases.** A quick cross-provider challenge here can prevent wasting an entire factory run on incomplete scenario coverage.\n\n**After orchestrate.sh parses the spec (Phase 1-2) and before embrace execution (Phase 4), dispatch a scenario coverage review:**\n\nIf a second provider is available, dispatch the challenge through Octopus routing:\n\n```bash\n# Read the spec to extract behaviors and constraints\nSPEC_CONTENT=$(\u003C\"\u003Cspec_path>\")\nreview_provider=\"\"\ncommand -v codex >\u002Fdev\u002Fnull 2>&1 && review_provider=\"codex\"\n[[ -z \"$review_provider\" ]] && command -v agy >\u002Fdev\u002Fnull 2>&1 && review_provider=\"agy\"\n\nif [[ -n \"$review_provider\" ]]; then\n  \"${HOME}\u002F.claude-octopus\u002Fplugin\u002Fscripts\u002Forchestrate.sh\" spawn \"$review_provider\" \\\n    \"You are a QA adversary. Given this specification and generated scenarios, identify coverage gaps.\n\nSPECIFICATION:\n${SPEC_CONTENT}\n\nAnswer:\n1. Which spec BEHAVIORS have no scenario testing them?\n2. Which EDGE CASES from the spec constraints are untested?\n3. Which FAILURE MODES are not covered (network failures, invalid input, concurrent access)?\n4. Rate overall coverage: SUFFICIENT \u002F GAPS-FOUND \u002F CRITICAL-GAPS\n\nBe specific — cite the behavior or constraint ID that lacks coverage.\" 2>\u002Fdev\u002Fnull || true\nfi\n```\n\n**After receiving the challenge response:**\n- If CRITICAL-GAPS found: warn the user and suggest refining the spec with `\u002Focto:spec` before proceeding\n- If GAPS-FOUND: note the gaps but proceed — the holdout phase (Phase 5) will catch some of these\n- If SUFFICIENT: proceed with confidence\n\n**This is a lightweight gate — it adds ~30 seconds but can save a $2.00 factory run on a spec with poor scenario coverage.**\n\n**Skip with `--fast` or when user explicitly requests speed over thoroughness.**\n\n---\n\n### STEP 5: Execute orchestrate.sh factory (MANDATORY — Bash Tool)\n```bash\n${HOME}\u002F.claude-octopus\u002Fplugin\u002Fscripts\u002Forchestrate.sh factory --spec \"\u003Cspec_path>\"\n```\n\nWith optional flags based on Step 1 answers:\n- `--holdout-ratio \u003Cvalue>` if custom ratio requested\n- `--max-retries \u003Cvalue>` if custom retry count\n- `--ci` if user approved non-interactive mode\n\n\u003CHARD-GATE>\nPROHIBITED from:\n- Running embrace directly (MUST use factory command which wraps it)\n- Simulating or faking holdout testing\n- Substituting direct Claude analysis for multi-provider scoring\n- Skipping the factory pipeline\n- Creating working\u002Fprogress files in plugin directory\n\u003C\u002FHARD-GATE>\n\n### STEP 6: Verify Factory Report (MANDATORY — Validation Gate)\n```bash\nREPORT_FILE=$(find .octo\u002Ffactory -name \"factory-report.md\" -mmin -60 2>\u002Fdev\u002Fnull | sort -r | head -1)\nif [[ -z \"$REPORT_FILE\" ]]; then\n    echo \"ERROR: Factory report not found\"\n    exit 1\nfi\ncat \"$REPORT_FILE\"\n```\nIf validation fails: report error, show logs from `~\u002F.claude-octopus\u002Flogs\u002F`, DO NOT proceed, DO NOT substitute.\n\n### STEP 7: Read Scores and Present Results (MANDATORY)\n```bash\nSCORES_FILE=$(find .octo\u002Ffactory -name \"satisfaction-scores.json\" -mmin -60 2>\u002Fdev\u002Fnull | sort -r | head -1)\nif [[ -n \"$SCORES_FILE\" ]]; then\n    cat \"$SCORES_FILE\"\nfi\n```\n\nPresent to user:\n1. **Verdict** with emoji (PASS\u002FWARN\u002FFAIL)\n2. **Composite score** vs satisfaction target\n3. **Dimension breakdown** (behavior, constraints, holdout, quality)\n4. **Holdout highlights** — which blind scenarios passed\u002Ffailed\n5. **Artifact directory** path for deep review\n\n### STEP 8: Update State & Next Steps (MANDATORY)\n```bash\n\"${HOME}\u002F.claude-octopus\u002Fplugin\u002Fscripts\u002Fstate-manager.sh\" record_decision \"factory\" \"Factory run completed: \u003Cverdict> (\u003Cscore>\u002F\u003Ctarget>)\" 2>\u002Fdev\u002Fnull || true\n```\n\nPresent next-step suggestions based on verdict:\n- **PASS:** Implementation meets spec. Review artifacts, run manual testing, ship.\n- **WARN:** Close to target. Review holdout failures, consider targeted fixes.\n- **FAIL:** Below target. Review `holdout-results.md` for specific failures. Consider:\n  - Refining the NLSpec with `\u002Focto:spec`\n  - Manual fixes + re-run with `--max-retries 2`\n  - Breaking spec into smaller, clearer pieces\n\nInclude attribution footer:\n```\nDark Factory Mode powered by Claude Octopus v8.25.0\nPipeline: Spec → Scenarios → Embrace → Holdout → Score → Report\nProviders: Codex | Antigravity | Claude\n```\n\n## Error Handling (by step)\n- Step 1 (Questions): If user declines all, proceed with defaults\n- Step 2 (Providers): Both external unavailable → Claude-only mode (warn)\n- Step 3 (Spec): File missing → STOP. Thin spec → WARN and proceed.\n- Step 4 (State): Failure → continue without state\n- Step 5 (orchestrate.sh): Show bash error, check logs — DO NOT substitute\n- Step 6 (Report): Missing → show logs, DO NOT proceed\n- Step 7 (Scores): Missing JSON → extract from report markdown\n- Step 8 (State): Failure → skip state update, still present results\n\n## Prohibited Actions\n- CANNOT skip orchestrate.sh factory execution\n- CANNOT simulate or fake the factory pipeline\n- CANNOT substitute direct Claude analysis for multi-provider scoring\n- CANNOT skip spec validation gate\n- CANNOT proceed past a failed validation gate\n- CANNOT create working\u002Fprogress files in plugin directory\n",{"name":5,"aliases":24,"description":7,"execution_mode":29,"validation_gates":30,"disable-model-invocation":34},[25,26,27,28],"factory","dark-factory","build-from-spec","autonomous-build","enforced",[31,32,33],"spec_file_validated","orchestrate_sh_executed","factory_report_exists",true,"main","d69096e239f2a89aabc2cd078c47664a9d3b1e11","05-技能与代理开发\u002Fskill-factory-claude-octopus",3,"https:\u002F\u002Fgithub.com\u002Flza6\u002FArmor-piercing-system-prompts","https:\u002F\u002Fgithub.com\u002Flza6","个人vx：Tf00798\r\n个人✈：@TFGY999","2026-10-08 14:29:22",{"source":8,"skillName":5,"commands":44,"prompt":54,"zipAvailable":34},[45,48,51],{"tool":46,"command":47},"npx","npx skills add lza6\u002FArmor-piercing-system-prompts --skill skill-factory",{"tool":49,"command":50},"bunx","bunx skills add lza6\u002FArmor-piercing-system-prompts --skill skill-factory",{"tool":52,"command":53},"pnpm","pnpm dlx skills add lza6\u002FArmor-piercing-system-prompts --skill skill-factory","安装技能，技能地址：https:\u002F\u002Fgithub.com\u002Flza6\u002FArmor-piercing-system-prompts\u002Fblob\u002Fmain\u002F05-技能与代理开发\u002Fskill-factory-claude-octopus\u002FSKILL.md",[],[57,64,69,76,83,90],{"skillNo":58,"name":59,"nameZh":6,"description":60,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":61,"categories":62,"keywords":63,"publishedAt":19},"01M7FFG485KY83TD6HSC4YPCMC","obsidian-clipper-template-creator","Guide for creating templates for the Obsidian Web Clipper. Use when you want to create a new clipping template, understand available variables, or format clipped content.","https:\u002F\u002Fgithub.com\u002Flza6\u002FArmor-piercing-system-prompts\u002Fblob\u002Fmain\u002F07-后端与架构\u002Fobsidian-clipper-template-creator\u002FSKILL.md",[],[],{"skillNo":65,"name":59,"nameZh":6,"description":60,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":66,"categories":67,"keywords":68,"publishedAt":19},"01M7FFG0T6AZ0STEJPSB3CX84K","https:\u002F\u002Fgithub.com\u002Flza6\u002FArmor-piercing-system-prompts\u002Fblob\u002Fmain\u002F07-后端与架构\u002Fobsidian-clipper-template-creator-antigravity-awesome-skills-main\u002FSKILL.md",[],[],{"skillNo":70,"name":71,"nameZh":6,"description":72,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":73,"categories":74,"keywords":75,"publishedAt":19},"01M7FFFY0QEM14SKWPDYGMXM2P","python-api","Programmatic Object-Detection-Metrics API guidance for BoundingBox, BoundingBoxes, Evaluator, enums, coordinate conversion, AP\u002FmAP semantics, and source-style import diagnostics.","https:\u002F\u002Fgithub.com\u002Flza6\u002FArmor-piercing-system-prompts\u002Fblob\u002Fmain\u002F07-后端与架构\u002Fobject-detection-metrics\u002Fsub-skills\u002Fpython-api\u002FSKILL.md",[],[],{"skillNo":77,"name":78,"nameZh":6,"description":79,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":80,"categories":81,"keywords":82,"publishedAt":19},"01M7FFFVZQZ2GJ0C0TGP7JHD20","file-evaluation","Folder-based PASCAL VOC AP\u002FmAP evaluation from ground-truth and detection text files.","https:\u002F\u002Fgithub.com\u002Flza6\u002FArmor-piercing-system-prompts\u002Fblob\u002Fmain\u002F07-后端与架构\u002Fobject-detection-metrics\u002Fsub-skills\u002Ffile-evaluation\u002FSKILL.md",[],[],{"skillNo":84,"name":85,"nameZh":6,"description":86,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":87,"categories":88,"keywords":89,"publishedAt":19},"01M7FFFSFNYSE77SJ6PE1CK2DA","object-detection-metrics","Guides legacy Object-Detection-Metrics PASCAL VOC AP\u002FmAP evaluation from detection text folders and direct Python API classes.","https:\u002F\u002Fgithub.com\u002Flza6\u002FArmor-piercing-system-prompts\u002Fblob\u002Fmain\u002F07-后端与架构\u002Fobject-detection-metrics\u002FSKILL.md",[],[],{"skillNo":91,"name":92,"nameZh":6,"description":93,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":94,"categories":95,"keywords":96,"publishedAt":19},"01M7FFFP0GCK1RVXBCPQ8APBMT","nx-workspace-patterns","Configure and optimize Nx monorepo workspaces. Use when setting up Nx, configuring project boundaries, optimizing build caching, or implementing affected commands.","https:\u002F\u002Fgithub.com\u002Flza6\u002FArmor-piercing-system-prompts\u002Fblob\u002Fmain\u002F07-后端与架构\u002Fnx-workspace-patterns\u002FSKILL.md",[],[]]