[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"skill:01M6C5BDF77QABH18J7C46VAN9":3,"skill-same:01M6C5BDF77QABH18J7C46VAN9":96},{"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":38,"linkOnly":39,"markdownContent":40,"markdownBody":41,"frontMatter":42,"refName":51,"commitSha":52,"skillDirectory":53,"forksCount":14,"repositoryHtmlUrl":54,"repositoryHomepageUrl":6,"ownerHtmlUrl":55,"ownerBlogUrl":56,"ownerDescription":6,"updatedAt":57,"install":58,"resources":71},"01M6C5BDF77QABH18J7C46VAN9","semgrep",null,"Run Semgrep static analysis scan on a codebase using parallel subagents. Supports two scan modes — \"run all\" (full ruleset coverage) and \"important only\" (high-confidence security vulnerabilities). Automatically detects and uses Semgrep Pro for cross-file taint analysis when available. Use when asked to scan code for vulnerabilities, run a security audit with Semgrep, find bugs, or perform static analysis. Spawns parallel workers for multi-language codebases.","CoralShades\u002Facm-ai","CoralShades","https:\u002F\u002Favatars.githubusercontent.com\u002Fu\u002F220297585?v=4","01M6C5211QSJHZ8HXSDDWCR507","Coral Shades AI Agency","01M6C52161FNQDRQ0FRR8DAA9D",0,"MIT","https:\u002F\u002Fgithub.com\u002FCoralShades\u002Facm-ai\u002Fblob\u002Fmain\u002F.agents\u002Fskills\u002Fsemgrep\u002FSKILL.md",[],[5,19,20,21,22,23,24,25,26,27,28,29,30,31,32,33,34,35,36,37],"security audit","only","use","subagents","coverage","read","find","parallel","Bash","find bugs","security","analysis","audit","file","code","run","scan code","vulnerabilities","static analysis","2026-11-10 10:12:30",false,"---\nname: semgrep\ndescription: >-\n  Run Semgrep static analysis scan on a codebase using parallel subagents.\n  Supports two scan modes — \"run all\" (full ruleset coverage) and \"important\n  only\" (high-confidence security vulnerabilities). Automatically detects and\n  uses Semgrep Pro for cross-file taint analysis when available. Use when asked\n  to scan code for vulnerabilities, run a security audit with Semgrep, find\n  bugs, or perform static analysis. Spawns parallel workers for multi-language\n  codebases.\nallowed-tools:\n  - Bash\n  - Read\n  - Glob\n  - Task\n  - AskUserQuestion\n  - TaskCreate\n  - TaskList\n  - TaskUpdate\n---\n\n# Semgrep Security Scan\n\nRun a Semgrep scan with automatic language detection, parallel execution via Task subagents, and merged SARIF output.\n\n## Essential Principles\n\n1. **Always use `--metrics=off`** — Semgrep sends telemetry by default; `--config auto` also phones home. Every `semgrep` command must include `--metrics=off` to prevent data leakage during security audits.\n2. **User must approve the scan plan (Step 3 is a hard gate)** — The original \"scan this codebase\" request is NOT approval. Present exact rulesets, target, engine, and mode; wait for explicit \"yes\"\u002F\"proceed\" before spawning scanners.\n3. **Third-party rulesets are required, not optional** — Trail of Bits, 0xdea, and Decurity rules catch vulnerabilities absent from the official registry. Include them whenever the detected language matches.\n4. **Spawn all scan Tasks in a single message** — Parallel execution is the core performance advantage. Never spawn Tasks sequentially; always emit all Task tool calls in one response.\n5. **Always check for Semgrep Pro before scanning** — Pro enables cross-file taint tracking and catches ~250% more true positives. Skipping the check means silently missing critical inter-file vulnerabilities.\n\n## When to Use\n\n- Security audit of a codebase\n- Finding vulnerabilities before code review\n- Scanning for known bug patterns\n- First-pass static analysis\n\n## When NOT to Use\n\n- Binary analysis → Use binary analysis tools\n- Already have Semgrep CI configured → Use existing pipeline\n- Need cross-file analysis but no Pro license → Consider CodeQL as alternative\n- Creating custom Semgrep rules → Use `semgrep-rule-creator` skill\n- Porting existing rules to other languages → Use `semgrep-rule-variant-creator` skill\n\n## Output Directory\n\nAll scan results, SARIF files, and temporary data are stored in a single output directory.\n\n- **If the user specifies an output directory** in their prompt, use it as `OUTPUT_DIR`.\n- **If not specified**, default to `.\u002Fstatic_analysis_semgrep_1`. If that already exists, increment to `_2`, `_3`, etc.\n\nIn both cases, **always create the directory** with `mkdir -p` before writing any files.\n\n```bash\n# Resolve output directory\nif [ -n \"$USER_SPECIFIED_DIR\" ]; then\n  OUTPUT_DIR=\"$USER_SPECIFIED_DIR\"\nelse\n  BASE=\"static_analysis_semgrep\"\n  N=1\n  while [ -e \"${BASE}_${N}\" ]; do\n    N=$((N + 1))\n  done\n  OUTPUT_DIR=\"${BASE}_${N}\"\nfi\nmkdir -p \"$OUTPUT_DIR\u002Fraw\" \"$OUTPUT_DIR\u002Fresults\"\n```\n\nThe output directory is resolved **once** at the start of Step 1 and used throughout all subsequent steps.\n\n```\n$OUTPUT_DIR\u002F\n├── rulesets.txt                 # Approved rulesets (logged after Step 3)\n├── raw\u002F                         # Per-scan raw output (unfiltered)\n│   ├── python-python.json\n│   ├── python-python.sarif\n│   ├── python-django.json\n│   ├── python-django.sarif\n│   └── ...\n└── results\u002F                     # Final merged output\n    └── results.sarif\n```\n\n## Prerequisites\n\n**Required:** Semgrep CLI (`semgrep --version`). If not installed, see [Semgrep installation docs](https:\u002F\u002Fsemgrep.dev\u002Fdocs\u002Fgetting-started\u002F).\n\n**Optional:** Semgrep Pro — enables cross-file taint tracking, inter-procedural analysis, and additional languages (Apex, C#, Elixir). Check with:\n\n```bash\nsemgrep --pro --validate --config p\u002Fdefault 2>\u002Fdev\u002Fnull && echo \"Pro available\" || echo \"OSS only\"\n```\n\n**Limitations:** OSS mode cannot track data flow across files. Pro mode uses `-j 1` for cross-file analysis (slower per ruleset, but parallel rulesets compensate).\n\n## Scan Modes\n\nSelect mode in Step 2 of the workflow. Mode affects both scanner flags and post-processing.\n\n| Mode | Coverage | Findings Reported |\n|------|----------|-------------------|\n| **Run all** | All rulesets, all severity levels | Everything |\n| **Important only** | All rulesets, pre- and post-filtered | Security vulns only, medium-high confidence\u002Fimpact |\n\n**Important only** applies two filter layers:\n1. **Pre-filter**: `--severity MEDIUM --severity HIGH --severity CRITICAL` (CLI flag)\n2. **Post-filter**: JSON metadata — keeps only `category=security`, `confidence∈{MEDIUM,HIGH}`, `impact∈{MEDIUM,HIGH}`\n\nSee [scan-modes.md](references\u002Fscan-modes.md) for metadata criteria and jq filter commands.\n\n## Orchestration Architecture\n\n```\n┌──────────────────────────────────────────────────────────────────┐\n│ MAIN AGENT (this skill)                                          │\n│ Step 1: Detect languages + check Pro availability                │\n│ Step 2: Select scan mode + rulesets (ref: rulesets.md)           │\n│ Step 3: Present plan + rulesets, get approval [⛔ HARD GATE]     │\n│ Step 4: Spawn parallel scan Tasks (approved rulesets + mode)     │\n│ Step 5: Merge results and report                                 │\n└──────────────────────────────────────────────────────────────────┘\n         │ Step 4\n         ▼\n┌─────────────────┐\n│ Scan Tasks      │\n│ (parallel)      │\n├─────────────────┤\n│ Python scanner  │\n│ JS\u002FTS scanner   │\n│ Go scanner      │\n│ Docker scanner  │\n└─────────────────┘\n```\n\n## Workflow\n\n**Follow the detailed workflow in [scan-workflow.md](workflows\u002Fscan-workflow.md).** Summary:\n\n| Step | Action | Gate | Key Reference |\n|------|--------|------|---------------|\n| 1 | Resolve output dir, detect languages + Pro availability | — | Use Glob, not Bash |\n| 2 | Select scan mode + rulesets | — | [rulesets.md](references\u002Frulesets.md) |\n| 3 | Present plan, get explicit approval | ⛔ HARD | AskUserQuestion |\n| 4 | Spawn parallel scan Tasks | — | [scanner-task-prompt.md](references\u002Fscanner-task-prompt.md) |\n| 5 | Merge results and report | — | Merge script (below) |\n\n**Task enforcement:** On invocation, create 5 tasks with blockedBy dependencies (each step blocks the previous). Step 3 is a HARD GATE — mark complete ONLY after user explicitly approves.\n\n**Merge command (Step 5):**\n\n```bash\nuv run {baseDir}\u002Fscripts\u002Fmerge_sarif.py $OUTPUT_DIR\u002Fraw $OUTPUT_DIR\u002Fresults\u002Fresults.sarif\n```\n\n## Agents\n\n| Agent | Tools | Purpose |\n|-------|-------|---------|\n| `static-analysis:semgrep-scanner` | Bash | Executes parallel semgrep scans for a language category |\n\nUse `subagent_type: static-analysis:semgrep-scanner` in Step 4 when spawning Task subagents.\n\n## Rationalizations to Reject\n\n| Shortcut | Why It's Wrong |\n|----------|----------------|\n| \"User asked for scan, that's approval\" | Original request ≠ plan approval. Present plan, use AskUserQuestion, await explicit \"yes\" |\n| \"Step 3 task is blocking, just mark complete\" | Lying about task status defeats enforcement. Only mark complete after real approval |\n| \"I already know what they want\" | Assumptions cause scanning wrong directories\u002Frulesets. Present plan for verification |\n| \"Just use default rulesets\" | User must see and approve exact rulesets before scan |\n| \"Add extra rulesets without asking\" | Modifying approved list without consent breaks trust |\n| \"Third-party rulesets are optional\" | Trail of Bits, 0xdea, Decurity catch vulnerabilities not in official registry — REQUIRED |\n| \"Use --config auto\" | Sends metrics; less control over rulesets |\n| \"One Task at a time\" | Defeats parallelism; spawn all Tasks together |\n| \"Pro is too slow, skip --pro\" | Cross-file analysis catches 250% more true positives; worth the time |\n| \"Semgrep handles GitHub URLs natively\" | URL handling fails on repos with non-standard YAML; always clone first |\n| \"Cleanup is optional\" | Cloned repos pollute the user's workspace and accumulate across runs |\n| \"Use `.` or relative path as target\" | Subagents need absolute paths to avoid ambiguity |\n| \"Let the user pick an output dir later\" | Output directory must be resolved at Step 1, before any files are created |\n\n## Reference Index\n\n| File | Content |\n|------|---------|\n| [rulesets.md](references\u002Frulesets.md) | Complete ruleset catalog and selection algorithm |\n| [scan-modes.md](references\u002Fscan-modes.md) | Pre\u002Fpost-filter criteria and jq commands |\n| [scanner-task-prompt.md](references\u002Fscanner-task-prompt.md) | Template for spawning scanner subagents |\n\n| Workflow | Purpose |\n|----------|---------|\n| [scan-workflow.md](workflows\u002Fscan-workflow.md) | Complete 5-step scan execution process |\n\n## Success Criteria\n\n- [ ] Output directory resolved (user-specified or auto-incremented default)\n- [ ] All generated files stored inside `$OUTPUT_DIR`\n- [ ] Languages detected with file counts; Pro status checked\n- [ ] Scan mode selected by user (run all \u002F important only)\n- [ ] Rulesets include third-party rules for all detected languages\n- [ ] User explicitly approved the scan plan (Step 3 gate passed)\n- [ ] All scan Tasks spawned in a single message and completed\n- [ ] Every `semgrep` command used `--metrics=off`\n- [ ] Approved rulesets logged to `$OUTPUT_DIR\u002Frulesets.txt`\n- [ ] Raw per-scan outputs stored in `$OUTPUT_DIR\u002Fraw\u002F`\n- [ ] `results.sarif` exists in `$OUTPUT_DIR\u002Fresults\u002F` and is valid JSON\n- [ ] Important-only mode: post-filter applied before merge; unfiltered results preserved in `raw\u002F`\n- [ ] Results summary reported with severity and category breakdown\n- [ ] Cloned repos (if any) cleaned up from `$OUTPUT_DIR\u002Frepos\u002F`\n","# Semgrep Security Scan\n\nRun a Semgrep scan with automatic language detection, parallel execution via Task subagents, and merged SARIF output.\n\n## Essential Principles\n\n1. **Always use `--metrics=off`** — Semgrep sends telemetry by default; `--config auto` also phones home. Every `semgrep` command must include `--metrics=off` to prevent data leakage during security audits.\n2. **User must approve the scan plan (Step 3 is a hard gate)** — The original \"scan this codebase\" request is NOT approval. Present exact rulesets, target, engine, and mode; wait for explicit \"yes\"\u002F\"proceed\" before spawning scanners.\n3. **Third-party rulesets are required, not optional** — Trail of Bits, 0xdea, and Decurity rules catch vulnerabilities absent from the official registry. Include them whenever the detected language matches.\n4. **Spawn all scan Tasks in a single message** — Parallel execution is the core performance advantage. Never spawn Tasks sequentially; always emit all Task tool calls in one response.\n5. **Always check for Semgrep Pro before scanning** — Pro enables cross-file taint tracking and catches ~250% more true positives. Skipping the check means silently missing critical inter-file vulnerabilities.\n\n## When to Use\n\n- Security audit of a codebase\n- Finding vulnerabilities before code review\n- Scanning for known bug patterns\n- First-pass static analysis\n\n## When NOT to Use\n\n- Binary analysis → Use binary analysis tools\n- Already have Semgrep CI configured → Use existing pipeline\n- Need cross-file analysis but no Pro license → Consider CodeQL as alternative\n- Creating custom Semgrep rules → Use `semgrep-rule-creator` skill\n- Porting existing rules to other languages → Use `semgrep-rule-variant-creator` skill\n\n## Output Directory\n\nAll scan results, SARIF files, and temporary data are stored in a single output directory.\n\n- **If the user specifies an output directory** in their prompt, use it as `OUTPUT_DIR`.\n- **If not specified**, default to `.\u002Fstatic_analysis_semgrep_1`. If that already exists, increment to `_2`, `_3`, etc.\n\nIn both cases, **always create the directory** with `mkdir -p` before writing any files.\n\n```bash\n# Resolve output directory\nif [ -n \"$USER_SPECIFIED_DIR\" ]; then\n  OUTPUT_DIR=\"$USER_SPECIFIED_DIR\"\nelse\n  BASE=\"static_analysis_semgrep\"\n  N=1\n  while [ -e \"${BASE}_${N}\" ]; do\n    N=$((N + 1))\n  done\n  OUTPUT_DIR=\"${BASE}_${N}\"\nfi\nmkdir -p \"$OUTPUT_DIR\u002Fraw\" \"$OUTPUT_DIR\u002Fresults\"\n```\n\nThe output directory is resolved **once** at the start of Step 1 and used throughout all subsequent steps.\n\n```\n$OUTPUT_DIR\u002F\n├── rulesets.txt                 # Approved rulesets (logged after Step 3)\n├── raw\u002F                         # Per-scan raw output (unfiltered)\n│   ├── python-python.json\n│   ├── python-python.sarif\n│   ├── python-django.json\n│   ├── python-django.sarif\n│   └── ...\n└── results\u002F                     # Final merged output\n    └── results.sarif\n```\n\n## Prerequisites\n\n**Required:** Semgrep CLI (`semgrep --version`). If not installed, see [Semgrep installation docs](https:\u002F\u002Fsemgrep.dev\u002Fdocs\u002Fgetting-started\u002F).\n\n**Optional:** Semgrep Pro — enables cross-file taint tracking, inter-procedural analysis, and additional languages (Apex, C#, Elixir). Check with:\n\n```bash\nsemgrep --pro --validate --config p\u002Fdefault 2>\u002Fdev\u002Fnull && echo \"Pro available\" || echo \"OSS only\"\n```\n\n**Limitations:** OSS mode cannot track data flow across files. Pro mode uses `-j 1` for cross-file analysis (slower per ruleset, but parallel rulesets compensate).\n\n## Scan Modes\n\nSelect mode in Step 2 of the workflow. Mode affects both scanner flags and post-processing.\n\n| Mode | Coverage | Findings Reported |\n|------|----------|-------------------|\n| **Run all** | All rulesets, all severity levels | Everything |\n| **Important only** | All rulesets, pre- and post-filtered | Security vulns only, medium-high confidence\u002Fimpact |\n\n**Important only** applies two filter layers:\n1. **Pre-filter**: `--severity MEDIUM --severity HIGH --severity CRITICAL` (CLI flag)\n2. **Post-filter**: JSON metadata — keeps only `category=security`, `confidence∈{MEDIUM,HIGH}`, `impact∈{MEDIUM,HIGH}`\n\nSee [scan-modes.md](references\u002Fscan-modes.md) for metadata criteria and jq filter commands.\n\n## Orchestration Architecture\n\n```\n┌──────────────────────────────────────────────────────────────────┐\n│ MAIN AGENT (this skill)                                          │\n│ Step 1: Detect languages + check Pro availability                │\n│ Step 2: Select scan mode + rulesets (ref: rulesets.md)           │\n│ Step 3: Present plan + rulesets, get approval [⛔ HARD GATE]     │\n│ Step 4: Spawn parallel scan Tasks (approved rulesets + mode)     │\n│ Step 5: Merge results and report                                 │\n└──────────────────────────────────────────────────────────────────┘\n         │ Step 4\n         ▼\n┌─────────────────┐\n│ Scan Tasks      │\n│ (parallel)      │\n├─────────────────┤\n│ Python scanner  │\n│ JS\u002FTS scanner   │\n│ Go scanner      │\n│ Docker scanner  │\n└─────────────────┘\n```\n\n## Workflow\n\n**Follow the detailed workflow in [scan-workflow.md](workflows\u002Fscan-workflow.md).** Summary:\n\n| Step | Action | Gate | Key Reference |\n|------|--------|------|---------------|\n| 1 | Resolve output dir, detect languages + Pro availability | — | Use Glob, not Bash |\n| 2 | Select scan mode + rulesets | — | [rulesets.md](references\u002Frulesets.md) |\n| 3 | Present plan, get explicit approval | ⛔ HARD | AskUserQuestion |\n| 4 | Spawn parallel scan Tasks | — | [scanner-task-prompt.md](references\u002Fscanner-task-prompt.md) |\n| 5 | Merge results and report | — | Merge script (below) |\n\n**Task enforcement:** On invocation, create 5 tasks with blockedBy dependencies (each step blocks the previous). Step 3 is a HARD GATE — mark complete ONLY after user explicitly approves.\n\n**Merge command (Step 5):**\n\n```bash\nuv run {baseDir}\u002Fscripts\u002Fmerge_sarif.py $OUTPUT_DIR\u002Fraw $OUTPUT_DIR\u002Fresults\u002Fresults.sarif\n```\n\n## Agents\n\n| Agent | Tools | Purpose |\n|-------|-------|---------|\n| `static-analysis:semgrep-scanner` | Bash | Executes parallel semgrep scans for a language category |\n\nUse `subagent_type: static-analysis:semgrep-scanner` in Step 4 when spawning Task subagents.\n\n## Rationalizations to Reject\n\n| Shortcut | Why It's Wrong |\n|----------|----------------|\n| \"User asked for scan, that's approval\" | Original request ≠ plan approval. Present plan, use AskUserQuestion, await explicit \"yes\" |\n| \"Step 3 task is blocking, just mark complete\" | Lying about task status defeats enforcement. Only mark complete after real approval |\n| \"I already know what they want\" | Assumptions cause scanning wrong directories\u002Frulesets. Present plan for verification |\n| \"Just use default rulesets\" | User must see and approve exact rulesets before scan |\n| \"Add extra rulesets without asking\" | Modifying approved list without consent breaks trust |\n| \"Third-party rulesets are optional\" | Trail of Bits, 0xdea, Decurity catch vulnerabilities not in official registry — REQUIRED |\n| \"Use --config auto\" | Sends metrics; less control over rulesets |\n| \"One Task at a time\" | Defeats parallelism; spawn all Tasks together |\n| \"Pro is too slow, skip --pro\" | Cross-file analysis catches 250% more true positives; worth the time |\n| \"Semgrep handles GitHub URLs natively\" | URL handling fails on repos with non-standard YAML; always clone first |\n| \"Cleanup is optional\" | Cloned repos pollute the user's workspace and accumulate across runs |\n| \"Use `.` or relative path as target\" | Subagents need absolute paths to avoid ambiguity |\n| \"Let the user pick an output dir later\" | Output directory must be resolved at Step 1, before any files are created |\n\n## Reference Index\n\n| File | Content |\n|------|---------|\n| [rulesets.md](references\u002Frulesets.md) | Complete ruleset catalog and selection algorithm |\n| [scan-modes.md](references\u002Fscan-modes.md) | Pre\u002Fpost-filter criteria and jq commands |\n| [scanner-task-prompt.md](references\u002Fscanner-task-prompt.md) | Template for spawning scanner subagents |\n\n| Workflow | Purpose |\n|----------|---------|\n| [scan-workflow.md](workflows\u002Fscan-workflow.md) | Complete 5-step scan execution process |\n\n## Success Criteria\n\n- [ ] Output directory resolved (user-specified or auto-incremented default)\n- [ ] All generated files stored inside `$OUTPUT_DIR`\n- [ ] Languages detected with file counts; Pro status checked\n- [ ] Scan mode selected by user (run all \u002F important only)\n- [ ] Rulesets include third-party rules for all detected languages\n- [ ] User explicitly approved the scan plan (Step 3 gate passed)\n- [ ] All scan Tasks spawned in a single message and completed\n- [ ] Every `semgrep` command used `--metrics=off`\n- [ ] Approved rulesets logged to `$OUTPUT_DIR\u002Frulesets.txt`\n- [ ] Raw per-scan outputs stored in `$OUTPUT_DIR\u002Fraw\u002F`\n- [ ] `results.sarif` exists in `$OUTPUT_DIR\u002Fresults\u002F` and is valid JSON\n- [ ] Important-only mode: post-filter applied before merge; unfiltered results preserved in `raw\u002F`\n- [ ] Results summary reported with severity and category breakdown\n- [ ] Cloned repos (if any) cleaned up from `$OUTPUT_DIR\u002Frepos\u002F`\n",{"name":5,"description":7,"allowed-tools":43},[27,44,45,46,47,48,49,50],"Read","Glob","Task","AskUserQuestion","TaskCreate","TaskList","TaskUpdate","main","1ca81b0a7de199942400f965edd4d7544077ac5d",".agents\u002Fskills\u002Fsemgrep","https:\u002F\u002Fgithub.com\u002FCoralShades\u002Facm-ai","https:\u002F\u002Fgithub.com\u002FCoralShades","coralshades.com.au","2026-04-18 00:45:07",{"source":8,"skillName":5,"commands":59,"prompt":69,"zipAvailable":70},[60,63,66],{"tool":61,"command":62},"npx","npx skills add CoralShades\u002Facm-ai --skill semgrep",{"tool":64,"command":65},"bunx","bunx skills add CoralShades\u002Facm-ai --skill semgrep",{"tool":67,"command":68},"pnpm","pnpm dlx skills add CoralShades\u002Facm-ai --skill semgrep","安装技能，技能地址：https:\u002F\u002Fgithub.com\u002FCoralShades\u002Facm-ai\u002Fblob\u002Fmain\u002F.agents\u002Fskills\u002Fsemgrep\u002FSKILL.md",true,[72,78,82,86,92],{"path":73,"resourceType":74,"resourceTypeDesc":75,"sizeBytes":76,"sourceHtmlUrl":77},"references\u002Frulesets.md","REFERENCE","参考文档",7695,"https:\u002F\u002Fgithub.com\u002FCoralShades\u002Facm-ai\u002Fblob\u002Fmain\u002F.agents\u002Fskills\u002Fsemgrep\u002Freferences\u002Frulesets.md",{"path":79,"resourceType":74,"resourceTypeDesc":75,"sizeBytes":80,"sourceHtmlUrl":81},"references\u002Fscan-modes.md",4477,"https:\u002F\u002Fgithub.com\u002FCoralShades\u002Facm-ai\u002Fblob\u002Fmain\u002F.agents\u002Fskills\u002Fsemgrep\u002Freferences\u002Fscan-modes.md",{"path":83,"resourceType":74,"resourceTypeDesc":75,"sizeBytes":84,"sourceHtmlUrl":85},"references\u002Fscanner-task-prompt.md",5909,"https:\u002F\u002Fgithub.com\u002FCoralShades\u002Facm-ai\u002Fblob\u002Fmain\u002F.agents\u002Fskills\u002Fsemgrep\u002Freferences\u002Fscanner-task-prompt.md",{"path":87,"resourceType":88,"resourceTypeDesc":89,"sizeBytes":90,"sourceHtmlUrl":91},"scripts\u002Fmerge_sarif.py","SCRIPT","脚本",6631,"https:\u002F\u002Fgithub.com\u002FCoralShades\u002Facm-ai\u002Fblob\u002Fmain\u002F.agents\u002Fskills\u002Fsemgrep\u002Fscripts\u002Fmerge_sarif.py",{"path":93,"resourceType":74,"resourceTypeDesc":75,"sizeBytes":94,"sourceHtmlUrl":95},"workflows\u002Fscan-workflow.md",11430,"https:\u002F\u002Fgithub.com\u002FCoralShades\u002Facm-ai\u002Fblob\u002Fmain\u002F.agents\u002Fskills\u002Fsemgrep\u002Fworkflows\u002Fscan-workflow.md",[97,119,131,147,169,179],{"skillNo":98,"name":99,"nameZh":6,"description":100,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":101,"categories":102,"keywords":108,"publishedAt":38},"01M6C63HKXA890HHV47WBAX0TB","uncodixfy","Prevents generic AI\u002FCodex UI patterns when generating frontend code. Use this skill whenever generating HTML, CSS, React, Vue, Svelte, or any frontend UI code to enforce clean, human-designed aesthetics inspired by Linear, Raycast, Stripe, and GitHub instead of typical AI-generated UI.","https:\u002F\u002Fgithub.com\u002FCoralShades\u002Facm-ai\u002Fblob\u002Fmain\u002F.qwen\u002Fskills\u002Funcodixfy\u002FSKILL.md",[103],{"categoryNo":104,"categoryName":105,"taxonomyType":106,"taxonomyTypeDesc":107},"01JBSEEDCAT000000000000009","前端开发","SCENARIO","使用场景",[99,109,110,111,112,113,33,114,115,116,117,21,118],"github","frontend","css","codex","react","linear","ui","vue","generated","html",{"skillNo":120,"name":121,"nameZh":6,"description":122,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":123,"licenseFileUrl":6,"sourceHtmlUrl":124,"categories":125,"keywords":126,"publishedAt":38},"01M6C63FJ2GEBBGZ5XSMV40KXB","runpodctl","Runpod CLI to manage your GPU workloads.","Apache-2.0","https:\u002F\u002Fgithub.com\u002FCoralShades\u002Facm-ai\u002Fblob\u002Fmain\u002F.qwen\u002Fskills\u002Frunpodctl\u002FSKILL.md",[],[121,127,128,129,130,27],"macos","cli","linux","gpu",{"skillNo":132,"name":133,"nameZh":6,"description":134,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":135,"categories":136,"keywords":137,"publishedAt":38},"01M6C63DPAV46TFVQ01SFYPJ7X","langgraph-persistence","INVOKE THIS SKILL when your LangGraph needs to persist state, remember conversations, travel through history, or configure subgraph checkpointer scoping. Covers checkpointers, thread_id, time travel, Store, and subgraph persistence modes.","https:\u002F\u002Fgithub.com\u002FCoralShades\u002Facm-ai\u002Fblob\u002Fmain\u002F.qwen\u002Fskills\u002Flanggraph-persistence\u002FSKILL.md",[],[138,139,140,141,142,143,144,145,146],"langgraph persistence","langgraph","persistence","travel","history","scoping","remember","state","configure",{"skillNo":148,"name":149,"nameZh":6,"description":150,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":151,"categories":152,"keywords":158,"publishedAt":38},"01M6C63BNS9CQEBTB8RNSSBFPA","langgraph-human-in-the-loop","INVOKE THIS SKILL when implementing human-in-the-loop patterns, pausing for approval, or handling errors in LangGraph. Covers interrupt(), Command(resume=...), approval\u002Fvalidation workflows, and the 4-tier error handling strategy.","https:\u002F\u002Fgithub.com\u002FCoralShades\u002Facm-ai\u002Fblob\u002Fmain\u002F.qwen\u002Fskills\u002Flanggraph-human-in-the-loop\u002FSKILL.md",[153],{"categoryNo":154,"categoryName":155,"taxonomyType":156,"taxonomyTypeDesc":157},"01JBSEEDCAT000000000000038","人力资源","PROFESSION","适用职业",[159,160,139,161,162,163,164,165,166,167,168],"langgraph human in the loop","loop","workflows","error handling","validation","human-in-the-loop","command","strategy","approval","resume",{"skillNo":170,"name":171,"nameZh":6,"description":172,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":173,"categories":174,"keywords":175,"publishedAt":38},"01M6C639GVYZFX3NFPV5FBDR4F","langgraph-fundamentals","INVOKE THIS SKILL when writing ANY LangGraph code. Covers StateGraph, state schemas, nodes, edges, Command, Send, invoke, streaming, and error handling.","https:\u002F\u002Fgithub.com\u002FCoralShades\u002Facm-ai\u002Fblob\u002Fmain\u002F.qwen\u002Fskills\u002Flanggraph-fundamentals\u002FSKILL.md",[],[176,139,162,145,177,178,165,33],"langgraph fundamentals","streaming","writing",{"skillNo":180,"name":181,"nameZh":6,"description":182,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":183,"categories":184,"keywords":185,"publishedAt":38},"01M6C6375N34YX2GZ3SD3R7XR2","langchain-rag","INVOKE THIS SKILL when building ANY retrieval-augmented generation (RAG) system. Covers document loaders, RecursiveCharacterTextSplitter, embeddings (OpenAI), and vector stores (Chroma, FAISS, Pinecone).","https:\u002F\u002Fgithub.com\u002FCoralShades\u002Facm-ai\u002Fblob\u002Fmain\u002F.qwen\u002Fskills\u002Flangchain-rag\u002FSKILL.md",[],[186,187,188,189,190,191,192,193,194,195],"rag","langchain rag","langchain","embeddings","vector","openai","chroma","faiss","vector stores","system"]