[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"skill:01M6C5AD37CFQ7C53JXSTD2P6X":3,"skill-same:01M6C5AD37CFQ7C53JXSTD2P6X":97},{"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":23,"publishedAt":43,"linkOnly":44,"markdownContent":45,"markdownBody":46,"frontMatter":47,"refName":48,"commitSha":49,"skillDirectory":50,"forksCount":14,"repositoryHtmlUrl":51,"repositoryHomepageUrl":6,"ownerHtmlUrl":52,"ownerBlogUrl":53,"ownerDescription":6,"updatedAt":54,"install":55,"resources":68},"01M6C5AD37CFQ7C53JXSTD2P6X","prompt-generator",null,"Generate optimized Claude Code prompts for any request. This is the main entry point for the prompt generation system. Use this skill whenever you need to create a prompt for a Claude Code session, generate a session prompt pack, prepare a multi-session plan, or translate a natural language request into a structured, actionable Claude Code prompt. Trigger on: 'generate a prompt for', 'create a prompt', 'help me write a prompt', 'what prompt should I use', '\u002Fgenerate-prompt', any use of the generate-prompt slash command, or when you want to ensure a Claude Code session starts with the right skills, context, glossary, and verification checklist loaded. Also trigger when someone wants to create session prompts similar to the S4-S9 pipeline prompt packs.","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\u002Fprompt-generator\u002FSKILL.md",[18],{"categoryNo":19,"categoryName":20,"taxonomyType":21,"taxonomyTypeDesc":22},"01JBSEEDCAT000000000000023","Skill与智能体开发","SCENARIO","使用场景",[24,25,26,27,28,29,30,31,32,33,34,35,36,37,38,39,40,41,42],"prompt generator","prompt","system","prompts","use","session","context","write","command","glossary","translate","plan","pipeline","verification","help","code","generate","create","prepare","2026-11-10 10:12:30",false,"---\nname: prompt-generator\ndescription: \"Generate optimized Claude Code prompts for any request. This is the main entry point for the prompt generation system. Use this skill whenever you need to create a prompt for a Claude Code session, generate a session prompt pack, prepare a multi-session plan, or translate a natural language request into a structured, actionable Claude Code prompt. Trigger on: 'generate a prompt for', 'create a prompt', 'help me write a prompt', 'what prompt should I use', '\u002Fgenerate-prompt', any use of the generate-prompt slash command, or when you want to ensure a Claude Code session starts with the right skills, context, glossary, and verification checklist loaded. Also trigger when someone wants to create session prompts similar to the S4-S9 pipeline prompt packs.\"\n---\n\n# Prompt Generator\n\nThe capstone skill that orchestrates the full prompt generation pipeline: discovery → classification → routing → generation. Produces ready-to-use Claude Code session prompts with glossary, verification checklist, and optional plan scaffolding.\n\n---\n\n## Phase 1 — Discover\n\nCheck if the skills registry is current:\n\n```bash\nREGISTRY=\"D:\u002Failocal\u002Facm-ai\u002Fskills-registry.json\"\n\n# Check existence and age (regenerate if missing or >1 hour old)\nif [ ! -f \"$REGISTRY\" ]; then\n  echo \"Registry missing — running skill-discovery\"\n  RUN_DISCOVERY=true\nelse\n  # Check modification time (cross-platform: stat -c on Linux, stat -f on macOS)\n  MTIME=$(stat -c \"%Y\" \"$REGISTRY\" 2>\u002Fdev\u002Fnull || stat -f \"%m\" \"$REGISTRY\" 2>\u002Fdev\u002Fnull)\n  NOW=$(date +%s)\n  AGE=$((NOW - MTIME))\n  if [ $AGE -gt 3600 ]; then\n    echo \"Registry stale ($((AGE \u002F 60)) min old) — running skill-discovery\"\n    RUN_DISCOVERY=true\n  fi\nfi\n```\n\nIf `RUN_DISCOVERY=true`, invoke `\u002Fskill-discovery` to refresh the registry before proceeding.\n\nRead `D:\u002Failocal\u002Facm-ai\u002Fskills-registry.json` after confirming it is current.\n\n---\n\n## Phase 2 — Classify\n\nApply `\u002Frequest-classifier` to the user's request. Produce a `RequestClassification` JSON.\n\nIf the classification is **ambiguous** (two types score within 1 point of each other, or the request contains mixed signals), present the candidate classifications to the user and ask for confirmation before proceeding:\n\n```\nI see two plausible classifications for your request:\n  A) pipeline \u002F complexity 6 \u002F plan ON\n  B) improvement \u002F complexity 5 \u002F plan ON\n\nThe strategies diverge here — A routes to a tmux team, B routes to a solo agent with subagent gates.\nWhich fits better, or shall I go with A (pipeline) since there are LangGraph signals?\n```\n\n---\n\n## Phase 3 — Route\n\nApply `\u002Fprompt-router` with the confirmed classification + skills registry.\n\nProduce a `PromptPlan` JSON. If the plan is **complex** (agent_strategy=tmux-team OR plan_mode=true with 5+ skills), present it for user confirmation before generating:\n\n```\nRouting plan:\n  Type: pipeline \u002F complexity 7 \u002F plan mode ON\n  Skills: \u002Fplanning-with-files, \u002Flanggraph-fundamentals, \u002Facm-observability, \u002Fpydantic-models-py\n  Strategy: tmux-team (3 panes: orchestrator, backend-dev, verifier)\n  Context7: LangGraph docs + LangChain docs\n  Output: prompt-pack → docs\u002Fsprint-artifacts\u002Fprompt-packs\u002F\n\nProceed? [Y\u002Fn]\n```\n\n---\n\n## Phase 4 — Generate\n\n### 4a — Load Template\n\nRead the master template from:\n```\nD:\u002Failocal\u002Facm-ai\u002F.claude\u002Fskills\u002Fprompt-generator\u002Freferences\u002Fprompt-template.md\n```\n\n### 4b — Build Glossary\n\nRead the glossary builder:\n```\nD:\u002Failocal\u002Facm-ai\u002F.claude\u002Fskills\u002Fprompt-generator\u002Freferences\u002Fglossary-builder.md\n```\n\nSelect the domain(s) based on `domain_signals` from the classification:\n- `extraction`, `pipeline`, `graph`, `node`, `langgraph` → **Pipeline domain**\n- `component`, `page`, `UI`, `React`, `css`, `frontend` → **Frontend domain**\n- Always include → **General domain**\n\nCap at 15 entries. Prioritize terms that appear in the user's original request text.\n\n### 4c — Populate Template\n\nFill in all `{{ variable }}` placeholders:\n\n| Placeholder | Source |\n|---|---|\n| `{{ session_title }}` | One-sentence goal from user's request |\n| `{{ skill_directives }}` | `\u002Fskill-name` lines from `PromptPlan.selected_skills` |\n| `{{ prerequisites }}` | Services\u002Ffiles that must exist before the session starts |\n| `{{ glossary_table }}` | Built in step 4b |\n| `{{ current_state }}` | Key state facts relevant to the request (branch, last sprint, etc.) |\n| `{{ key_files_list }}` | Exact absolute paths for files the session will touch |\n| `{{ plan_or_steps }}` | If plan_mode: scaffold plan format; else: \"What to Change\" section |\n| `{{ strategy_config }}` | Agent strategy block from `PromptPlan.agent_config` |\n| `{{ context7_section }}` | Context7 directives from `PromptPlan.context7_directives` (omit if empty) |\n| `{{ verification_items }}` | Checklist items from `PromptPlan.verification_items` |\n| `{{ files_summary }}` | NEW \u002F MODIFY \u002F MOVE counts from key files list |\n| `{{ commit_message }}` | Conventional commit template (`feat:` \u002F `fix:` \u002F `refactor:`) |\n\n### 4d — Plan Mode Scaffolding\n\nIf `plan_mode=true`, also create these files in `docs\u002Fsprint-artifacts\u002F`:\n\n**task_plan.md** skeleton:\n```markdown\n# Task Plan: {session_title}\nDate: {YYYY-MM-DD}\nStatus: IN PROGRESS\n\n## Goal\n{one-sentence goal}\n\n## Steps\n- [ ] Step 1\n- [ ] Step 2\n...\n\n## Risks\n- (none identified yet)\n```\n\n**findings.md** skeleton:\n```markdown\n# Findings: {session_title}\nDate: {YYYY-MM-DD}\n\n## What Was Discovered\n(populate during session)\n\n## Decisions Made\n(populate during session)\n```\n\n**progress.md** skeleton:\n```markdown\n# Progress: {session_title}\nDate: {YYYY-MM-DD}\n\n## Completed\n(none yet)\n\n## In Progress\n(none yet)\n\n## Blocked\n(none yet)\n```\n\n---\n\n## Phase 5 — Output\n\nSelect output format from `PromptPlan.output_format`:\n\n### terminal (default)\nPrint directly with clear markers:\n```\n══════════════════════════════════════════\n  GENERATED PROMPT — {session_title}\n══════════════════════════════════════════\n{populated template content}\n══════════════════════════════════════════\n```\n\n### copy-paste\nPrint in a fenced code block so the user can copy it cleanly:\n````\n```prompt\n{populated template content}\n```\n````\n\n### prompt-pack\nSave to `docs\u002Fsprint-artifacts\u002Fprompt-packs\u002F{YYYY-MM-DD}-{slug}.md` where `{slug}` is the session title kebab-cased.\n\nPrint a confirmation:\n```\nPrompt pack saved to:\n  docs\u002Fsprint-artifacts\u002Fprompt-packs\u002F2026-03-13-fix-extraction-timeout.md\n```\n\n### --save flag override\nIf the user passed `--save`, always save to the prompt-pack path regardless of `output_format`, AND print to terminal.\n\n---\n\n## Quick Start\n\nThree example invocations:\n\n**1. Simple fix (terminal output, no plan):**\n```\n\u002Fgenerate-prompt \"Fix the building sidebar not loading when source has 0 buildings\"\n```\n→ Classifies as bug-fix\u002Fsimple, routes to solo+systematic-debugging, outputs inline.\n\n**2. Complex feature (saved prompt-pack + plan scaffolding):**\n```\n\u002Fgenerate-prompt \"Add MinerU as a new extraction provider with fallback chain\" --save --tmux\n```\n→ Classifies as pipeline\u002Fcomplex, routes to tmux-team+langgraph-fundamentals, saves to prompt-packs\u002F, creates task_plan.md.\n\n**3. Refactor with explicit format:**\n```\n\u002Fgenerate-prompt \"Refactor pre-extraction stages to reduce LLM calls\" --format prompt-pack\n```\n→ Classifies as improvement\u002Fmedium, routes to solo+verification-before-completion, saves as prompt-pack.\n\n---\n\n## Flags Reference\n\n| Flag | Effect |\n|---|---|\n| `--save` | Always save to prompt-packs\u002F, also print to terminal |\n| `--no-plan` | Force `plan_mode=false`, skip plan scaffolding |\n| `--with-plan` | Force `plan_mode=true`, always scaffold task_plan.md |\n| `--tmux` | Force `agent_strategy=tmux-team` regardless of classification |\n| `--format terminal` | Print with markers (default) |\n| `--format copy-paste` | Print in fenced code block |\n| `--format prompt-pack` | Save to file only |\n\n---\n\n## Notes\n\n- If `skills-registry.json` is missing, Phase 1 runs `\u002Fskill-discovery` automatically — this adds ~5 seconds but ensures accurate skill selection\n- Plan scaffolding files are created relative to the repo root; the session prompt references them with absolute paths\n- Context7 directives in the generated prompt are ready-to-execute — copy them verbatim at session start\n- The generated prompt is designed to be pasted as the first message of a new Claude Code session, not run in the current conversation\n","# Prompt Generator\n\nThe capstone skill that orchestrates the full prompt generation pipeline: discovery → classification → routing → generation. Produces ready-to-use Claude Code session prompts with glossary, verification checklist, and optional plan scaffolding.\n\n---\n\n## Phase 1 — Discover\n\nCheck if the skills registry is current:\n\n```bash\nREGISTRY=\"D:\u002Failocal\u002Facm-ai\u002Fskills-registry.json\"\n\n# Check existence and age (regenerate if missing or >1 hour old)\nif [ ! -f \"$REGISTRY\" ]; then\n  echo \"Registry missing — running skill-discovery\"\n  RUN_DISCOVERY=true\nelse\n  # Check modification time (cross-platform: stat -c on Linux, stat -f on macOS)\n  MTIME=$(stat -c \"%Y\" \"$REGISTRY\" 2>\u002Fdev\u002Fnull || stat -f \"%m\" \"$REGISTRY\" 2>\u002Fdev\u002Fnull)\n  NOW=$(date +%s)\n  AGE=$((NOW - MTIME))\n  if [ $AGE -gt 3600 ]; then\n    echo \"Registry stale ($((AGE \u002F 60)) min old) — running skill-discovery\"\n    RUN_DISCOVERY=true\n  fi\nfi\n```\n\nIf `RUN_DISCOVERY=true`, invoke `\u002Fskill-discovery` to refresh the registry before proceeding.\n\nRead `D:\u002Failocal\u002Facm-ai\u002Fskills-registry.json` after confirming it is current.\n\n---\n\n## Phase 2 — Classify\n\nApply `\u002Frequest-classifier` to the user's request. Produce a `RequestClassification` JSON.\n\nIf the classification is **ambiguous** (two types score within 1 point of each other, or the request contains mixed signals), present the candidate classifications to the user and ask for confirmation before proceeding:\n\n```\nI see two plausible classifications for your request:\n  A) pipeline \u002F complexity 6 \u002F plan ON\n  B) improvement \u002F complexity 5 \u002F plan ON\n\nThe strategies diverge here — A routes to a tmux team, B routes to a solo agent with subagent gates.\nWhich fits better, or shall I go with A (pipeline) since there are LangGraph signals?\n```\n\n---\n\n## Phase 3 — Route\n\nApply `\u002Fprompt-router` with the confirmed classification + skills registry.\n\nProduce a `PromptPlan` JSON. If the plan is **complex** (agent_strategy=tmux-team OR plan_mode=true with 5+ skills), present it for user confirmation before generating:\n\n```\nRouting plan:\n  Type: pipeline \u002F complexity 7 \u002F plan mode ON\n  Skills: \u002Fplanning-with-files, \u002Flanggraph-fundamentals, \u002Facm-observability, \u002Fpydantic-models-py\n  Strategy: tmux-team (3 panes: orchestrator, backend-dev, verifier)\n  Context7: LangGraph docs + LangChain docs\n  Output: prompt-pack → docs\u002Fsprint-artifacts\u002Fprompt-packs\u002F\n\nProceed? [Y\u002Fn]\n```\n\n---\n\n## Phase 4 — Generate\n\n### 4a — Load Template\n\nRead the master template from:\n```\nD:\u002Failocal\u002Facm-ai\u002F.claude\u002Fskills\u002Fprompt-generator\u002Freferences\u002Fprompt-template.md\n```\n\n### 4b — Build Glossary\n\nRead the glossary builder:\n```\nD:\u002Failocal\u002Facm-ai\u002F.claude\u002Fskills\u002Fprompt-generator\u002Freferences\u002Fglossary-builder.md\n```\n\nSelect the domain(s) based on `domain_signals` from the classification:\n- `extraction`, `pipeline`, `graph`, `node`, `langgraph` → **Pipeline domain**\n- `component`, `page`, `UI`, `React`, `css`, `frontend` → **Frontend domain**\n- Always include → **General domain**\n\nCap at 15 entries. Prioritize terms that appear in the user's original request text.\n\n### 4c — Populate Template\n\nFill in all `{{ variable }}` placeholders:\n\n| Placeholder | Source |\n|---|---|\n| `{{ session_title }}` | One-sentence goal from user's request |\n| `{{ skill_directives }}` | `\u002Fskill-name` lines from `PromptPlan.selected_skills` |\n| `{{ prerequisites }}` | Services\u002Ffiles that must exist before the session starts |\n| `{{ glossary_table }}` | Built in step 4b |\n| `{{ current_state }}` | Key state facts relevant to the request (branch, last sprint, etc.) |\n| `{{ key_files_list }}` | Exact absolute paths for files the session will touch |\n| `{{ plan_or_steps }}` | If plan_mode: scaffold plan format; else: \"What to Change\" section |\n| `{{ strategy_config }}` | Agent strategy block from `PromptPlan.agent_config` |\n| `{{ context7_section }}` | Context7 directives from `PromptPlan.context7_directives` (omit if empty) |\n| `{{ verification_items }}` | Checklist items from `PromptPlan.verification_items` |\n| `{{ files_summary }}` | NEW \u002F MODIFY \u002F MOVE counts from key files list |\n| `{{ commit_message }}` | Conventional commit template (`feat:` \u002F `fix:` \u002F `refactor:`) |\n\n### 4d — Plan Mode Scaffolding\n\nIf `plan_mode=true`, also create these files in `docs\u002Fsprint-artifacts\u002F`:\n\n**task_plan.md** skeleton:\n```markdown\n# Task Plan: {session_title}\nDate: {YYYY-MM-DD}\nStatus: IN PROGRESS\n\n## Goal\n{one-sentence goal}\n\n## Steps\n- [ ] Step 1\n- [ ] Step 2\n...\n\n## Risks\n- (none identified yet)\n```\n\n**findings.md** skeleton:\n```markdown\n# Findings: {session_title}\nDate: {YYYY-MM-DD}\n\n## What Was Discovered\n(populate during session)\n\n## Decisions Made\n(populate during session)\n```\n\n**progress.md** skeleton:\n```markdown\n# Progress: {session_title}\nDate: {YYYY-MM-DD}\n\n## Completed\n(none yet)\n\n## In Progress\n(none yet)\n\n## Blocked\n(none yet)\n```\n\n---\n\n## Phase 5 — Output\n\nSelect output format from `PromptPlan.output_format`:\n\n### terminal (default)\nPrint directly with clear markers:\n```\n══════════════════════════════════════════\n  GENERATED PROMPT — {session_title}\n══════════════════════════════════════════\n{populated template content}\n══════════════════════════════════════════\n```\n\n### copy-paste\nPrint in a fenced code block so the user can copy it cleanly:\n````\n```prompt\n{populated template content}\n```\n````\n\n### prompt-pack\nSave to `docs\u002Fsprint-artifacts\u002Fprompt-packs\u002F{YYYY-MM-DD}-{slug}.md` where `{slug}` is the session title kebab-cased.\n\nPrint a confirmation:\n```\nPrompt pack saved to:\n  docs\u002Fsprint-artifacts\u002Fprompt-packs\u002F2026-03-13-fix-extraction-timeout.md\n```\n\n### --save flag override\nIf the user passed `--save`, always save to the prompt-pack path regardless of `output_format`, AND print to terminal.\n\n---\n\n## Quick Start\n\nThree example invocations:\n\n**1. Simple fix (terminal output, no plan):**\n```\n\u002Fgenerate-prompt \"Fix the building sidebar not loading when source has 0 buildings\"\n```\n→ Classifies as bug-fix\u002Fsimple, routes to solo+systematic-debugging, outputs inline.\n\n**2. Complex feature (saved prompt-pack + plan scaffolding):**\n```\n\u002Fgenerate-prompt \"Add MinerU as a new extraction provider with fallback chain\" --save --tmux\n```\n→ Classifies as pipeline\u002Fcomplex, routes to tmux-team+langgraph-fundamentals, saves to prompt-packs\u002F, creates task_plan.md.\n\n**3. Refactor with explicit format:**\n```\n\u002Fgenerate-prompt \"Refactor pre-extraction stages to reduce LLM calls\" --format prompt-pack\n```\n→ Classifies as improvement\u002Fmedium, routes to solo+verification-before-completion, saves as prompt-pack.\n\n---\n\n## Flags Reference\n\n| Flag | Effect |\n|---|---|\n| `--save` | Always save to prompt-packs\u002F, also print to terminal |\n| `--no-plan` | Force `plan_mode=false`, skip plan scaffolding |\n| `--with-plan` | Force `plan_mode=true`, always scaffold task_plan.md |\n| `--tmux` | Force `agent_strategy=tmux-team` regardless of classification |\n| `--format terminal` | Print with markers (default) |\n| `--format copy-paste` | Print in fenced code block |\n| `--format prompt-pack` | Save to file only |\n\n---\n\n## Notes\n\n- If `skills-registry.json` is missing, Phase 1 runs `\u002Fskill-discovery` automatically — this adds ~5 seconds but ensures accurate skill selection\n- Plan scaffolding files are created relative to the repo root; the session prompt references them with absolute paths\n- Context7 directives in the generated prompt are ready-to-execute — copy them verbatim at session start\n- The generated prompt is designed to be pasted as the first message of a new Claude Code session, not run in the current conversation\n",{"name":5,"description":7},"main","1ca81b0a7de199942400f965edd4d7544077ac5d",".agents\u002Fskills\u002Fprompt-generator","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":56,"prompt":66,"zipAvailable":67},[57,60,63],{"tool":58,"command":59},"npx","npx skills add CoralShades\u002Facm-ai --skill prompt-generator",{"tool":61,"command":62},"bunx","bunx skills add CoralShades\u002Facm-ai --skill prompt-generator",{"tool":64,"command":65},"pnpm","pnpm dlx skills add CoralShades\u002Facm-ai --skill prompt-generator","安装技能，技能地址：https:\u002F\u002Fgithub.com\u002FCoralShades\u002Facm-ai\u002Fblob\u002Fmain\u002F.agents\u002Fskills\u002Fprompt-generator\u002FSKILL.md",true,[69,75,79,83,87,91],{"path":70,"resourceType":71,"resourceTypeDesc":72,"sizeBytes":73,"sourceHtmlUrl":74},"agents\u002Fopenai.yaml","REFERENCE","参考文档",250,"https:\u002F\u002Fgithub.com\u002FTimothyMeadows\u002FHephaera\u002Fblob\u002Fmain\u002Fskills\u002Frig-and-animate-blender-actors\u002Fagents\u002Fopenai.yaml",{"path":76,"resourceType":71,"resourceTypeDesc":72,"sizeBytes":77,"sourceHtmlUrl":78},"references\u002Fglossary-builder.md",9606,"https:\u002F\u002Fgithub.com\u002FCoralShades\u002Facm-ai\u002Fblob\u002Fmain\u002F.agents\u002Fskills\u002Fprompt-generator\u002Freferences\u002Fglossary-builder.md",{"path":80,"resourceType":71,"resourceTypeDesc":72,"sizeBytes":81,"sourceHtmlUrl":82},"references\u002Fprompt-template.md",6996,"https:\u002F\u002Fgithub.com\u002FCoralShades\u002Facm-ai\u002Fblob\u002Fmain\u002F.agents\u002Fskills\u002Fprompt-generator\u002Freferences\u002Fprompt-template.md",{"path":84,"resourceType":71,"resourceTypeDesc":72,"sizeBytes":85,"sourceHtmlUrl":86},"references\u002Frig-animation-authoring-contract.md",2119,"https:\u002F\u002Fgithub.com\u002FTimothyMeadows\u002FHephaera\u002Fblob\u002Fmain\u002Fskills\u002Frig-and-animate-blender-actors\u002Freferences\u002Frig-animation-authoring-contract.md",{"path":88,"resourceType":71,"resourceTypeDesc":72,"sizeBytes":89,"sourceHtmlUrl":90},"references\u002Fsubject-articulation-and-deformation.md",6581,"https:\u002F\u002Fgithub.com\u002FTimothyMeadows\u002FHephaera\u002Fblob\u002Fmain\u002Fskills\u002Frig-and-animate-blender-actors\u002Freferences\u002Fsubject-articulation-and-deformation.md",{"path":92,"resourceType":93,"resourceTypeDesc":94,"sizeBytes":95,"sourceHtmlUrl":96},"scripts\u002Fgenerate_prompt.sh","SCRIPT","脚本",10938,"https:\u002F\u002Fgithub.com\u002FCoralShades\u002Facm-ai\u002Fblob\u002Fmain\u002F.agents\u002Fskills\u002Fprompt-generator\u002Fscripts\u002Fgenerate_prompt.sh",[98,118,131,147,168,178],{"skillNo":99,"name":100,"nameZh":6,"description":101,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":102,"categories":103,"keywords":107,"publishedAt":43},"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",[104],{"categoryNo":105,"categoryName":106,"taxonomyType":21,"taxonomyTypeDesc":22},"01JBSEEDCAT000000000000009","前端开发",[100,108,109,110,111,112,39,113,114,115,116,28,117],"github","frontend","css","codex","react","linear","ui","vue","generated","html",{"skillNo":119,"name":120,"nameZh":6,"description":121,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":122,"licenseFileUrl":6,"sourceHtmlUrl":123,"categories":124,"keywords":125,"publishedAt":43},"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",[],[120,126,127,128,129,130],"macos","cli","linux","gpu","Bash",{"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":43},"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":43},"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",[159,160,139,161,162,163,164,32,165,166,167],"langgraph human in the loop","loop","workflows","error handling","validation","human-in-the-loop","strategy","approval","resume",{"skillNo":169,"name":170,"nameZh":6,"description":171,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":172,"categories":173,"keywords":174,"publishedAt":43},"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",[],[175,139,162,145,176,177,32,39],"langgraph fundamentals","streaming","writing",{"skillNo":179,"name":180,"nameZh":6,"description":181,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":182,"categories":183,"keywords":184,"publishedAt":43},"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",[],[185,186,187,188,189,190,191,192,193,26],"rag","langchain rag","langchain","embeddings","vector","openai","chroma","faiss","vector stores"]