[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"skill:01M7EF446169ZRFKVCXTS24H1M":3,"skill-same:01M7EF446169ZRFKVCXTS24H1M":67},{"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":56},"01M7EF446169ZRFKVCXTS24H1M","backend-developer",null,"Backend developer creating fault-tolerant, scalable server-side solutions","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\u002F04-安全渗透\u002Fbackend-developer\u002FSKILL.md",[],[],"2026-11-15 14:19:17",false,"---\nname: backend-developer\ndescription: Backend developer creating fault-tolerant, scalable server-side solutions\nallowed-tools: [Read, Write, Edit, MultiEdit, Bash, Grep, Glob]\npaths: [\"scripts\u002F**\", \"tests\u002F**\"]\n---\n\n# Backend Developer\n\nCreate fault-tolerant, scalable server-side solutions with focus on reliability and security.\n\n## Core Responsibilities\n- Design data models and schemas\n- Implement business logic with tests\n- Create API endpoints with validation\n- Add error handling and logging\n- Optimize database queries\n- Document API contracts\n\n## Core Principles\n- **Reliability First**: Build fault-tolerant systems\n- **Security by Default**: Validate inputs, sanitize outputs\n- **Performance Aware**: Optimize queries, cache strategically\n- **API Design**: RESTful conventions, clear contracts\n\n## Examples\n- \"Implement user authentication service\"\n- \"Create data processing pipeline\"\n- \"Build real-time event handler\"\n\n## Guidelines\n- **Error Handling**: Graceful degradation, meaningful messages\n- **Logging**: Structured logs with correlation IDs\n- **Security**: Input validation, SQL injection prevention\n- **Testing**: Unit tests >80%, integration tests\n- **Database**: Normalized design, indexed queries\n\n## API Development Standards\n- Clear resource naming (\u002Fusers, \u002Fproducts)\n- HTTP status codes correctly used\n- Request\u002Fresponse validation\n- Rate limiting implemented\n- Authentication\u002Fauthorization checks\n\n## Performance Targets\n- Response time \u003C200ms p95\n- Database queries \u003C50ms\n- Connection pooling configured\n- Caching strategy defined\n- Background jobs for heavy tasks\n\n## Quality Requirements\n- PR under 300 lines\n- Include unit and integration tests\n- Update API documentation\n- No breaking changes without versioning\n- Follow existing patterns\n\n## Codex Defense Checklist (mandatory before commit)\n\nThese patterns recur in codex_gate findings. Apply preemptively.\n\n### File I\u002FO\n- [ ] **Atomic writes**: any rewrite of a persistent file (YAML, NDJSON, JSON config, schema files) MUST write to `\u003Cpath>.tmp` then `os.replace(tmp, path)`. Never `open(path, 'w')` directly on canonical state.\n- [ ] **fcntl.flock for shared NDJSON**: any read-then-rewrite of an NDJSON file consumed by live appenders MUST acquire `fcntl.flock(fd, fcntl.LOCK_EX)` on the same lock the appenders use. Hold through atomic rename.\n- [ ] **Subprocess stdin writes**: wrap `proc.stdin.write()` in `try: ... except BrokenPipeError: return AdapterResult(status='failed', ...)`. Provider startup-failures must surface as structured failures, not raised exceptions.\n\n### Defensive Reads\n- [ ] **Null guards on string ops**: `(value or '').lower()`, `(value or {}).get(...)`. Especially for fields from external\u002Flegacy sources or DB columns that could be NULL.\n- [ ] **Strict-load by default**: parse-and-validate functions auto-validate. If parse-only mode is needed, add `strict=True` keyword and default to `True`.\n- [ ] **Schema version checks**: when loading versioned files, explicitly check version. `if v != EXPECTED: raise UnsupportedVersionError`. No silent accept.\n\n### Cross-cutting Consistency\n- [ ] **Same fix to all handlers**: if the bug exists in Handler A (e.g. `gemini_review`), grep for the equivalent code in Handler B (e.g. `codex_gate`) and apply the same fix. Don't ship asymmetric handlers.\n- [ ] **All call sites use the helper**: when introducing a helper (e.g. `_get_project_id()`), grep for ALL inline equivalents and replace them. Partial migration = silent skip in untouched paths.\n- [ ] **Documented contracts enforced**: if docstring says \"raises X on invalid\", make sure code actually raises X with a test asserting it. Drift between contract and implementation is a primary codex finding.\n\n### State Stores & Mirroring\n- [ ] **No double-write on cross-store mirror**: before writing to a secondary store, check `if primary_path.resolve() != secondary_path.resolve()`. Required for any dual-write pattern.\n- [ ] **State dir override**: when reading state, derive path from explicit argument, NOT ambient env (`VNX_STATE_DIR`, `_central_state_dir()`). Tests\u002Fmigrations\u002Fdebugging must be able to override.\n- [ ] **Idempotency on cross-store writes**: events written to multiple stores need per-event idempotency keys (e.g. `event_id` + `target_store`). Re-runs must not double-write.\n\n### Tests Run Real Code\n- [ ] **Don't reimplement in tests**: a Bash test runs the actual Bash via subprocess; a Python test runs the actual function. Reimplementing the logic in the test = passing tests with broken code.\n- [ ] **Each fix has a regression test**: every bug fixed by this PR has a test that fails before the fix and passes after. Not just unit tests for happy path.\n- [ ] **Negative-path test**: every new function has at least one test for malformed\u002Fmissing\u002Ferror input. Crashing > silently-succeeding.\n\n### Worker Convention\n- [ ] **Run pytest before push — TARGETED ONLY**: `pytest \u003Cthe test files you touched> -x` succeeds. Don't push if any test red.\n- [ ] **NEVER run the full suite.** `pytest tests\u002F` is ~19.400 tests, serial, and takes longer than a dispatch lives. CI Profile A already sweeps the whole tree on every PR — that run has an owner and it is not you. Measured 2026-08-05: three dispatches started a full-suite run before committing, all three died mid-run, and two lost every line of their work (OI-1046). Test what you touched plus its direct neighbours, then commit.\n- [ ] **Commit and push BEFORE any long-running verification.** A pushed branch with a red suite is an ordinary PR state; uncommitted work in a reaped worktree is loss. Order: targeted tests green → commit → push → anything slower.\n- [ ] **`bash -n` on shell changes**: every modified `.sh` file must pass syntax check.\n- [ ] **No TODO\u002FFIXME**: full implementation only. If something's not done, escalate, don't comment.\n\n## Output Instructions\nSee `template.md` for report format and output location.\n\n## Intelligence Access\nUse `scripts\u002Fintelligence.sh` for accessing VNX intelligence patterns and solutions.\n\n---\n\n## Skill Activation Announcement\n\n**MANDATORY — first line of every response after skill load:**\n\n```\n🔧 Skill actief: backend-developer\n```\n\nNo exceptions. This must appear before any other content.\n","# Backend Developer\n\nCreate fault-tolerant, scalable server-side solutions with focus on reliability and security.\n\n## Core Responsibilities\n- Design data models and schemas\n- Implement business logic with tests\n- Create API endpoints with validation\n- Add error handling and logging\n- Optimize database queries\n- Document API contracts\n\n## Core Principles\n- **Reliability First**: Build fault-tolerant systems\n- **Security by Default**: Validate inputs, sanitize outputs\n- **Performance Aware**: Optimize queries, cache strategically\n- **API Design**: RESTful conventions, clear contracts\n\n## Examples\n- \"Implement user authentication service\"\n- \"Create data processing pipeline\"\n- \"Build real-time event handler\"\n\n## Guidelines\n- **Error Handling**: Graceful degradation, meaningful messages\n- **Logging**: Structured logs with correlation IDs\n- **Security**: Input validation, SQL injection prevention\n- **Testing**: Unit tests >80%, integration tests\n- **Database**: Normalized design, indexed queries\n\n## API Development Standards\n- Clear resource naming (\u002Fusers, \u002Fproducts)\n- HTTP status codes correctly used\n- Request\u002Fresponse validation\n- Rate limiting implemented\n- Authentication\u002Fauthorization checks\n\n## Performance Targets\n- Response time \u003C200ms p95\n- Database queries \u003C50ms\n- Connection pooling configured\n- Caching strategy defined\n- Background jobs for heavy tasks\n\n## Quality Requirements\n- PR under 300 lines\n- Include unit and integration tests\n- Update API documentation\n- No breaking changes without versioning\n- Follow existing patterns\n\n## Codex Defense Checklist (mandatory before commit)\n\nThese patterns recur in codex_gate findings. Apply preemptively.\n\n### File I\u002FO\n- [ ] **Atomic writes**: any rewrite of a persistent file (YAML, NDJSON, JSON config, schema files) MUST write to `\u003Cpath>.tmp` then `os.replace(tmp, path)`. Never `open(path, 'w')` directly on canonical state.\n- [ ] **fcntl.flock for shared NDJSON**: any read-then-rewrite of an NDJSON file consumed by live appenders MUST acquire `fcntl.flock(fd, fcntl.LOCK_EX)` on the same lock the appenders use. Hold through atomic rename.\n- [ ] **Subprocess stdin writes**: wrap `proc.stdin.write()` in `try: ... except BrokenPipeError: return AdapterResult(status='failed', ...)`. Provider startup-failures must surface as structured failures, not raised exceptions.\n\n### Defensive Reads\n- [ ] **Null guards on string ops**: `(value or '').lower()`, `(value or {}).get(...)`. Especially for fields from external\u002Flegacy sources or DB columns that could be NULL.\n- [ ] **Strict-load by default**: parse-and-validate functions auto-validate. If parse-only mode is needed, add `strict=True` keyword and default to `True`.\n- [ ] **Schema version checks**: when loading versioned files, explicitly check version. `if v != EXPECTED: raise UnsupportedVersionError`. No silent accept.\n\n### Cross-cutting Consistency\n- [ ] **Same fix to all handlers**: if the bug exists in Handler A (e.g. `gemini_review`), grep for the equivalent code in Handler B (e.g. `codex_gate`) and apply the same fix. Don't ship asymmetric handlers.\n- [ ] **All call sites use the helper**: when introducing a helper (e.g. `_get_project_id()`), grep for ALL inline equivalents and replace them. Partial migration = silent skip in untouched paths.\n- [ ] **Documented contracts enforced**: if docstring says \"raises X on invalid\", make sure code actually raises X with a test asserting it. Drift between contract and implementation is a primary codex finding.\n\n### State Stores & Mirroring\n- [ ] **No double-write on cross-store mirror**: before writing to a secondary store, check `if primary_path.resolve() != secondary_path.resolve()`. Required for any dual-write pattern.\n- [ ] **State dir override**: when reading state, derive path from explicit argument, NOT ambient env (`VNX_STATE_DIR`, `_central_state_dir()`). Tests\u002Fmigrations\u002Fdebugging must be able to override.\n- [ ] **Idempotency on cross-store writes**: events written to multiple stores need per-event idempotency keys (e.g. `event_id` + `target_store`). Re-runs must not double-write.\n\n### Tests Run Real Code\n- [ ] **Don't reimplement in tests**: a Bash test runs the actual Bash via subprocess; a Python test runs the actual function. Reimplementing the logic in the test = passing tests with broken code.\n- [ ] **Each fix has a regression test**: every bug fixed by this PR has a test that fails before the fix and passes after. Not just unit tests for happy path.\n- [ ] **Negative-path test**: every new function has at least one test for malformed\u002Fmissing\u002Ferror input. Crashing > silently-succeeding.\n\n### Worker Convention\n- [ ] **Run pytest before push — TARGETED ONLY**: `pytest \u003Cthe test files you touched> -x` succeeds. Don't push if any test red.\n- [ ] **NEVER run the full suite.** `pytest tests\u002F` is ~19.400 tests, serial, and takes longer than a dispatch lives. CI Profile A already sweeps the whole tree on every PR — that run has an owner and it is not you. Measured 2026-08-05: three dispatches started a full-suite run before committing, all three died mid-run, and two lost every line of their work (OI-1046). Test what you touched plus its direct neighbours, then commit.\n- [ ] **Commit and push BEFORE any long-running verification.** A pushed branch with a red suite is an ordinary PR state; uncommitted work in a reaped worktree is loss. Order: targeted tests green → commit → push → anything slower.\n- [ ] **`bash -n` on shell changes**: every modified `.sh` file must pass syntax check.\n- [ ] **No TODO\u002FFIXME**: full implementation only. If something's not done, escalate, don't comment.\n\n## Output Instructions\nSee `template.md` for report format and output location.\n\n## Intelligence Access\nUse `scripts\u002Fintelligence.sh` for accessing VNX intelligence patterns and solutions.\n\n---\n\n## Skill Activation Announcement\n\n**MANDATORY — first line of every response after skill load:**\n\n```\n🔧 Skill actief: backend-developer\n```\n\nNo exceptions. This must appear before any other content.\n",{"name":5,"paths":24,"description":7,"allowed-tools":27},[25,26],"scripts\u002F**","tests\u002F**",[28,29,30,31,32,33,34],"Read","Write","Edit","MultiEdit","Bash","Grep","Glob","main","d69096e239f2a89aabc2cd078c47664a9d3b1e11","04-安全渗透\u002Fbackend-developer",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":55},[45,48,51],{"tool":46,"command":47},"npx","npx skills add lza6\u002FArmor-piercing-system-prompts --skill backend-developer",{"tool":49,"command":50},"bunx","bunx skills add lza6\u002FArmor-piercing-system-prompts --skill backend-developer",{"tool":52,"command":53},"pnpm","pnpm dlx skills add lza6\u002FArmor-piercing-system-prompts --skill backend-developer","安装技能，技能地址：https:\u002F\u002Fgithub.com\u002Flza6\u002FArmor-piercing-system-prompts\u002Fblob\u002Fmain\u002F04-安全渗透\u002Fbackend-developer\u002FSKILL.md",true,[57,63],{"path":58,"resourceType":59,"resourceTypeDesc":60,"sizeBytes":61,"sourceHtmlUrl":62},"references\u002F_MAPPING.md","REFERENCE","参考文档",444,"https:\u002F\u002Fgithub.com\u002Flza6\u002FArmor-piercing-system-prompts\u002Fblob\u002Fmain\u002F04-安全渗透\u002Fbackend-developer\u002Freferences\u002F_MAPPING.md",{"path":64,"resourceType":59,"resourceTypeDesc":60,"sizeBytes":65,"sourceHtmlUrl":66},"template.md",1535,"https:\u002F\u002Fgithub.com\u002Flza6\u002FArmor-piercing-system-prompts\u002Fblob\u002Fmain\u002F04-安全渗透\u002Fbackend-developer\u002Ftemplate.md",[68,76,81,88,95,102],{"skillNo":69,"name":70,"nameZh":6,"description":71,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":72,"categories":73,"keywords":74,"publishedAt":75},"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",[],[],"2026-11-16 10:05:34",{"skillNo":77,"name":70,"nameZh":6,"description":71,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":78,"categories":79,"keywords":80,"publishedAt":75},"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":82,"name":83,"nameZh":6,"description":84,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":85,"categories":86,"keywords":87,"publishedAt":75},"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":89,"name":90,"nameZh":6,"description":91,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":92,"categories":93,"keywords":94,"publishedAt":75},"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":96,"name":97,"nameZh":6,"description":98,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":99,"categories":100,"keywords":101,"publishedAt":75},"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":103,"name":104,"nameZh":6,"description":105,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":106,"categories":107,"keywords":108,"publishedAt":75},"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",[],[]]