[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"skill:01M6CYTM6GAZ5T6YA06V4PAC5J":3,"skill-same:01M6CYTM6GAZ5T6YA06V4PAC5J":86},{"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":28,"publishedAt":47,"linkOnly":48,"markdownContent":49,"markdownBody":50,"frontMatter":51,"refName":62,"commitSha":63,"skillDirectory":64,"forksCount":65,"repositoryHtmlUrl":66,"repositoryHomepageUrl":67,"ownerHtmlUrl":68,"ownerBlogUrl":69,"ownerDescription":70,"updatedAt":71,"install":72,"resources":85},"01M6CYTM6GAZ5T6YA06V4PAC5J","rlm",null,"Process large codebases (>100 files) using the Recursive Language Model pattern. Treats code as an external environment, using parallel background agents to map-reduce complex tasks without context rot.","eric861129\u002FSKILLS_All-in-one","eric861129","https:\u002F\u002Favatars.githubusercontent.com\u002Fu\u002F74950944?v=4","01M6CXHDYEYRPFGGV4EFXCSYCF","ChiYu-Huang","01M6CXHE2J5CN1EBTY0FKV3ZKW",52,"MIT","https:\u002F\u002Fgithub.com\u002Feric861129\u002FSKILLS_All-in-one\u002Fblob\u002Fmain\u002Fpublic\u002FSKILLS\u002FUtility & Automation\u002Fclawfu-mcp-skills\u002FSKILL.md",[18,23],{"categoryNo":19,"categoryName":20,"taxonomyType":21,"taxonomyTypeDesc":22},"01JBSEEDCAT000000000000023","Skill与智能体开发","SCENARIO","使用场景",{"categoryNo":24,"categoryName":25,"taxonomyType":26,"taxonomyTypeDesc":27},"01JBSEEDCAT000000000000030","开发工程师","PROFESSION","适用职业",[5,29,30,31,32,33,34,35,36,37,38,39,40,41,42,43,44,45,46],"environment","project","across","pattern","map","scan","language","files","process","MCP","context","find","parallel","model","analyze","repository","code","tasks","2026-11-10 10:12:30",false,"---\nname: rlm\ndescription: Process large codebases (>100 files) using the Recursive Language Model pattern. Treats code as an external environment, using parallel background agents to map-reduce complex tasks without context rot.\ntriggers:\n  - \"analyze codebase\"\n  - \"scan all files\"\n  - \"large repository\"\n  - \"RLM\"\n  - \"find usage of X across the project\"\nlicense: MIT\nmetadata:\n  author: ClawFu\n  version: 1.0.0\n  mcp-server: \"@clawfu\u002Fmcp-skills\"\n---\n\n# Recursive Language Model (RLM) Skill\n\n## Core Philosophy\n**\"Context is an external resource, not a local variable.\"**\n\nWhen this skill is active, you are the **Root Node** of a Recursive Language Model system. Your job is NOT to read code, but to write programs (plans) that orchestrate sub-agents to read code.\n\n## Protocol: The RLM Loop\n\n### Phase 1: Choose Your Engine\nDecide based on the nature of the data:\n\n| Engine | Use Case | Tool |\n|--------|----------|------|\n| **Native Mode** | General codebase traversal, finding files, structure. | `find`, `grep`, `bash` |\n| **Strict Mode** | Dense data analysis (logs, CSVs, massive single files). | `python3 ~\u002F.claude\u002Fskills\u002Frlm\u002Frlm.py` |\n\n### Phase 2: Index & Filter (The \"Peeking\" Phase)\n**Goal**: Identify relevant data without loading it.\n1.  **Native**: Use `find` or `grep -l`.\n2.  **Strict**: Use `python3 ...\u002Frlm.py peek \"query\"`.\n    *   *RLM Pattern*: Grepping for import statements, class names, or definitions to build a list of relevant paths.\n\n### Phase 3: Parallel Map (The \"Sub-Query\" Phase)\n**Goal**: Process chunks in parallel using fresh contexts.\n1.  **Divide**: Split the work into atomic units.\n    - **Strict Mode**: `python3 ...\u002Frlm.py chunk --pattern \"*.log\"` -> Returns JSON chunks.\n2.  **Spawn**: Use `background_task` to launch parallel agents.\n    *   *Constraint*: Launch at least 3-5 agents in parallel for broad tasks.\n    *   *Prompting*: Give each background agent ONE specific chunk or file path.\n    *   *Format*: `background_task(agent=\"explore\", prompt=\"Analyze chunk #5 of big.log: {content}...\")`\n\n### Phase 4: Reduce & Synthesize (The \"Aggregation\" Phase)\n**Goal**: Combine results into a coherent answer.\n1.  **Collect**: Read the outputs from `background_task` (via `background_output`).\n2.  **Synthesize**: Look for patterns, consensus, or specific answers in the aggregated data.\n3.  **Refine**: If the answer is incomplete, perform a second RLM recursion on the specific missing pieces.\n\n## Critical Instructions\n\n1.  **NEVER** use `cat *` or read more than 3-5 files into your main context at once.\n2.  **ALWAYS** prefer `background_task` for reading\u002Fanalyzing file contents when the file count > 1.\n3.  **Use `rlm.py`** for programmatic slicing of large files that `grep` can't handle well.\n4.  **Python is your Memory**: If you need to track state across 50 files, write a Python script (or use `rlm.py`) to scan them and output a summary.\n\n## Example Workflow: \"Find all API endpoints and check for Auth\"\n\n**Wrong Way (Monolithic)**:\n- `read src\u002Fapi\u002Froutes.ts`\n- `read src\u002Fapi\u002Fusers.ts`\n- ... (Context fills up, reasoning degrades)\n\n**RLM Way (Recursive)**:\n1.  **Filter**: `grep -l \"@Controller\" src\u002F**\u002F*.ts` -> Returns 20 files.\n2.  **Map**: \n    - `background_task(prompt=\"Read src\u002Fapi\u002Froutes.ts. Extract all endpoints and their @Auth decorators.\")`\n    - `background_task(prompt=\"Read src\u002Fapi\u002Fusers.ts. Extract all endpoints and their @Auth decorators.\")`\n    - ... (Launch all 20)\n3.  **Reduce**: \n    - Collect all 20 outputs.\n    - Compile into a single table.\n    - Identify missing auth.\n\n## Recovery Mode\nIf `background_task` is unavailable or fails:\n1.  Fall back to **Iterative Python Scripting**.\n2.  Write a Python script that loads each file, runs a regex\u002FAST check, and prints the result to stdout.\n3.  Read the script's stdout.\n\n---\n\n## What Claude Does vs What You Decide\n\n| Claude handles | You provide |\n|---------------|-------------|\n| Orchestrating parallel agents | Initial query and success criteria |\n| Chunking large files for processing | Judgment on result quality |\n| Synthesizing results from subagents | Final interpretation and action |\n| Writing filtering scripts | Validation of completeness |\n| Managing context isolation | Decision on when to stop recursing |\n\n---\n\n## Skill Boundaries\n\n### This skill excels for:\n- Codebases with >100 files\n- Finding patterns across many files\n- Audit tasks (security, auth, logging)\n- Large file analysis (logs, data dumps)\n\n### This skill is NOT ideal for:\n- Small projects (\u003C50 files) → Direct reading faster\n- Single file analysis → Overkill\n- Tasks requiring file modification → Use different approach\n\n---\n\n## Skill Metadata\n\n```yaml\nname: rlm\ncategory: meta\nversion: 2.0\nauthor: GUIA\nsource_expert: Recursive Language Model pattern\ndifficulty: advanced\nmode: cyborg\ntags: [rlm, large-codebase, parallel-agents, map-reduce, context-management]\ncreated: 2026-02-03\nupdated: 2026-02-03\n```\n","# Recursive Language Model (RLM) Skill\n\n## Core Philosophy\n**\"Context is an external resource, not a local variable.\"**\n\nWhen this skill is active, you are the **Root Node** of a Recursive Language Model system. Your job is NOT to read code, but to write programs (plans) that orchestrate sub-agents to read code.\n\n## Protocol: The RLM Loop\n\n### Phase 1: Choose Your Engine\nDecide based on the nature of the data:\n\n| Engine | Use Case | Tool |\n|--------|----------|------|\n| **Native Mode** | General codebase traversal, finding files, structure. | `find`, `grep`, `bash` |\n| **Strict Mode** | Dense data analysis (logs, CSVs, massive single files). | `python3 ~\u002F.claude\u002Fskills\u002Frlm\u002Frlm.py` |\n\n### Phase 2: Index & Filter (The \"Peeking\" Phase)\n**Goal**: Identify relevant data without loading it.\n1.  **Native**: Use `find` or `grep -l`.\n2.  **Strict**: Use `python3 ...\u002Frlm.py peek \"query\"`.\n    *   *RLM Pattern*: Grepping for import statements, class names, or definitions to build a list of relevant paths.\n\n### Phase 3: Parallel Map (The \"Sub-Query\" Phase)\n**Goal**: Process chunks in parallel using fresh contexts.\n1.  **Divide**: Split the work into atomic units.\n    - **Strict Mode**: `python3 ...\u002Frlm.py chunk --pattern \"*.log\"` -> Returns JSON chunks.\n2.  **Spawn**: Use `background_task` to launch parallel agents.\n    *   *Constraint*: Launch at least 3-5 agents in parallel for broad tasks.\n    *   *Prompting*: Give each background agent ONE specific chunk or file path.\n    *   *Format*: `background_task(agent=\"explore\", prompt=\"Analyze chunk #5 of big.log: {content}...\")`\n\n### Phase 4: Reduce & Synthesize (The \"Aggregation\" Phase)\n**Goal**: Combine results into a coherent answer.\n1.  **Collect**: Read the outputs from `background_task` (via `background_output`).\n2.  **Synthesize**: Look for patterns, consensus, or specific answers in the aggregated data.\n3.  **Refine**: If the answer is incomplete, perform a second RLM recursion on the specific missing pieces.\n\n## Critical Instructions\n\n1.  **NEVER** use `cat *` or read more than 3-5 files into your main context at once.\n2.  **ALWAYS** prefer `background_task` for reading\u002Fanalyzing file contents when the file count > 1.\n3.  **Use `rlm.py`** for programmatic slicing of large files that `grep` can't handle well.\n4.  **Python is your Memory**: If you need to track state across 50 files, write a Python script (or use `rlm.py`) to scan them and output a summary.\n\n## Example Workflow: \"Find all API endpoints and check for Auth\"\n\n**Wrong Way (Monolithic)**:\n- `read src\u002Fapi\u002Froutes.ts`\n- `read src\u002Fapi\u002Fusers.ts`\n- ... (Context fills up, reasoning degrades)\n\n**RLM Way (Recursive)**:\n1.  **Filter**: `grep -l \"@Controller\" src\u002F**\u002F*.ts` -> Returns 20 files.\n2.  **Map**: \n    - `background_task(prompt=\"Read src\u002Fapi\u002Froutes.ts. Extract all endpoints and their @Auth decorators.\")`\n    - `background_task(prompt=\"Read src\u002Fapi\u002Fusers.ts. Extract all endpoints and their @Auth decorators.\")`\n    - ... (Launch all 20)\n3.  **Reduce**: \n    - Collect all 20 outputs.\n    - Compile into a single table.\n    - Identify missing auth.\n\n## Recovery Mode\nIf `background_task` is unavailable or fails:\n1.  Fall back to **Iterative Python Scripting**.\n2.  Write a Python script that loads each file, runs a regex\u002FAST check, and prints the result to stdout.\n3.  Read the script's stdout.\n\n---\n\n## What Claude Does vs What You Decide\n\n| Claude handles | You provide |\n|---------------|-------------|\n| Orchestrating parallel agents | Initial query and success criteria |\n| Chunking large files for processing | Judgment on result quality |\n| Synthesizing results from subagents | Final interpretation and action |\n| Writing filtering scripts | Validation of completeness |\n| Managing context isolation | Decision on when to stop recursing |\n\n---\n\n## Skill Boundaries\n\n### This skill excels for:\n- Codebases with >100 files\n- Finding patterns across many files\n- Audit tasks (security, auth, logging)\n- Large file analysis (logs, data dumps)\n\n### This skill is NOT ideal for:\n- Small projects (\u003C50 files) → Direct reading faster\n- Single file analysis → Overkill\n- Tasks requiring file modification → Use different approach\n\n---\n\n## Skill Metadata\n\n```yaml\nname: rlm\ncategory: meta\nversion: 2.0\nauthor: GUIA\nsource_expert: Recursive Language Model pattern\ndifficulty: advanced\nmode: cyborg\ntags: [rlm, large-codebase, parallel-agents, map-reduce, context-management]\ncreated: 2026-02-03\nupdated: 2026-02-03\n```\n",{"name":5,"license":15,"metadata":52,"triggers":56,"description":7},{"author":53,"version":54,"mcp-server":55},"ClawFu","1.0.0","@clawfu\u002Fmcp-skills",[57,58,59,60,61],"analyze codebase","scan all files","large repository","RLM","find usage of X across the project","main","cbdd44aae21131521c09e42762e36178c790eacf","public\u002FSKILLS\u002FUtility & Automation\u002Fclawfu-mcp-skills",13,"https:\u002F\u002Fgithub.com\u002Feric861129\u002FSKILLS_All-in-one","http:\u002F\u002Fhuangchiyu.com\u002FSKILLS_All-in-one\u002F","https:\u002F\u002Fgithub.com\u002Feric861129","https:\u002F\u002Fhuangchiyu.com\u002FChiYu-Blog\u002F","開發工程師，專注於 C#／.NET 系統現代化、AI-assisted engineering、開源工具與技術寫作。","2026-04-16 14:28:14",{"source":8,"skillName":5,"commands":73,"prompt":83,"zipAvailable":84},[74,77,80],{"tool":75,"command":76},"npx","npx skills add eric861129\u002FSKILLS_All-in-one --skill rlm",{"tool":78,"command":79},"bunx","bunx skills add eric861129\u002FSKILLS_All-in-one --skill rlm",{"tool":81,"command":82},"pnpm","pnpm dlx skills add eric861129\u002FSKILLS_All-in-one --skill rlm","安装技能，技能地址：https:\u002F\u002Fgithub.com\u002Feric861129\u002FSKILLS_All-in-one\u002Fblob\u002Fmain\u002Fpublic\u002FSKILLS\u002FUtility & Automation\u002Fclawfu-mcp-skills\u002FSKILL.md",true,[],[87,115,146,171,196,216],{"skillNo":88,"name":89,"nameZh":6,"description":90,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":91,"categories":92,"keywords":94,"publishedAt":47},"01M6CZ1WR3KNVTKSWH4GNWT646","what-if-oracle","Run structured What-If scenario analysis with multi-branch possibility exploration. Use this skill when the user asks speculative questions like \"what if...\", \"what would happen if...\", \"what are the possibilities\", \"explore scenarios\", \"scenario analysis\", \"possibility space\", \"what could go wrong\", \"best case \u002F worst case\", \"risk analysis\", \"contingency planning\", \"strategic options\", or any question about uncertain futures. Also trigger when the user faces a fork-in-the-road decision, wants to stress-test an idea, or needs to think through consequences before committing.","https:\u002F\u002Fgithub.com\u002Feric861129\u002FSKILLS_All-in-one\u002Fblob\u002Fmain\u002Fpublic\u002FSKILLS\u002FWriting & Research\u002Fwhat-if-oracle\u002FSKILL.md",[93],{"categoryNo":19,"categoryName":20,"taxonomyType":21,"taxonomyTypeDesc":22},[95,96,97,98,99,100,101,102,103,104,105,106,107,108,109,110,111,112,113,114],"oracle","what if oracle","go","run","planning","ux","futures","risk","analysis","explore","decision","branch","question","read","write","idea","exploration","options","stress-test","strategies",{"skillNo":116,"name":117,"nameZh":6,"description":118,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":119,"categories":120,"keywords":131,"publishedAt":47},"01M6CZ1QHGJE91A95MSN5EK5Q9","venue-templates","Access comprehensive LaTeX templates, formatting requirements, and submission guidelines for major scientific publication venues (Nature, Science, PLOS, IEEE, ACM), academic conferences (NeurIPS, ICML, CVPR, CHI), research posters, and grant proposals (NSF, NIH, DOE, DARPA). This skill should be used when preparing manuscripts for journal submission, conference papers, research posters, or grant proposals and need venue-specific formatting requirements and templates.","https:\u002F\u002Fgithub.com\u002Feric861129\u002FSKILLS_All-in-one\u002Fblob\u002Fmain\u002Fpublic\u002FSKILLS\u002FWriting & Research\u002Fvenue-templates\u002FSKILL.md",[121,124,127,128],{"categoryNo":122,"categoryName":123,"taxonomyType":21,"taxonomyTypeDesc":22},"01JBSEEDCAT000000000000018","写作与文案",{"categoryNo":125,"categoryName":126,"taxonomyType":21,"taxonomyTypeDesc":22},"01JBSEEDCAT000000000000021","资料检索与研究",{"categoryNo":19,"categoryName":20,"taxonomyType":21,"taxonomyTypeDesc":22},{"categoryNo":129,"categoryName":130,"taxonomyType":26,"taxonomyTypeDesc":27},"01JBSEEDCAT000000000000036","内容创作者",[132,133,134,135,136,137,138,139,140,141,108,109,142,143,144,145],"templates","venue templates","research","guidelines","NeurIPS","ICML","LaTeX","science","Bash","access","edit","proposals","Academic","requirements",{"skillNo":147,"name":148,"nameZh":6,"description":149,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":150,"categories":151,"keywords":154,"publishedAt":47},"01M6CZ1KBQD2E5RTFB3BGVX7CD","the-fool","Use when challenging ideas, plans, decisions, or proposals using structured critical reasoning. Invoke to play devil's advocate, run a pre-mortem, red team, or audit evidence and assumptions.","https:\u002F\u002Fgithub.com\u002Feric861129\u002FSKILLS_All-in-one\u002Fblob\u002Fmain\u002Fpublic\u002FSKILLS\u002FWriting & Research\u002Fthe-fool\u002FSKILL.md",[152,153],{"categoryNo":19,"categoryName":20,"taxonomyType":21,"taxonomyTypeDesc":22},{"categoryNo":24,"categoryName":25,"taxonomyType":26,"taxonomyTypeDesc":27},[155,156,97,157,158,98,45,159,160,161,162,163,164,165,143,166,167,168,169,170],"the fool","review","workflow","architecture","audit","forge","devil","team","designer","reviewer","challenge","stress test","evidence","expert","poke","ideas",{"skillNo":172,"name":173,"nameZh":6,"description":174,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":175,"categories":176,"keywords":180,"publishedAt":47},"01M6CZ1ES826S7P1X3333E40TA","scientific-writing","Core skill for the deep research and writing tool. Write scientific manuscripts in full paragraphs (never bullet points). Use two-stage process with (1) section outlines with key points using research-lookup then (2) convert to flowing prose. IMRAD structure, citations (APA\u002FAMA\u002FVancouver), figures\u002Ftables, reporting guidelines (CONSORT\u002FSTROBE\u002FPRISMA), for research papers and journal submissions.","https:\u002F\u002Fgithub.com\u002Feric861129\u002FSKILLS_All-in-one\u002Fblob\u002Fmain\u002Fpublic\u002FSKILLS\u002FWriting & Research\u002Fscientific-writing\u002FSKILL.md",[177,178,179],{"categoryNo":122,"categoryName":123,"taxonomyType":21,"taxonomyTypeDesc":22},{"categoryNo":19,"categoryName":20,"taxonomyType":21,"taxonomyTypeDesc":22},{"categoryNo":129,"categoryName":130,"taxonomyType":26,"taxonomyTypeDesc":27},[181,182,135,183,184,185,186,187,140,188,108,109,142,189,190,191,192,193,194,195],"writing","scientific writing","core","reporting","deep research","Citations","Outlines","tool","use","never","structure","deep","full","stage","prisma",{"skillNo":197,"name":198,"nameZh":6,"description":199,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":200,"categories":201,"keywords":203,"publishedAt":47},"01M6CZ1B2JGJTB9FRD0TW4RXXE","scientific-brainstorming","Creative research ideation and exploration. Use for open-ended brainstorming sessions, exploring interdisciplinary connections, challenging assumptions, or identifying research gaps. Best for early-stage research planning when you do not have specific observations yet. For formulating testable hypotheses from data use hypothesis-generation.","https:\u002F\u002Fgithub.com\u002Feric861129\u002FSKILLS_All-in-one\u002Fblob\u002Fmain\u002Fpublic\u002FSKILLS\u002FWriting & Research\u002Fscientific-brainstorming\u002FSKILL.md",[202],{"categoryNo":19,"categoryName":20,"taxonomyType":21,"taxonomyTypeDesc":22},[204,205,134,206,99,207,208,209,189,210,211,212,111,213,214,194,215],"brainstorming","scientific brainstorming","sessions","data","hypothesis","open","ideation","creative","do","connections","research ideation","generation",{"skillNo":217,"name":218,"nameZh":6,"description":219,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":220,"categories":221,"keywords":228,"publishedAt":47},"01M6CZ16JG8N7A2ZX04BGE8J2Y","research-lookup","Look up current research information using the Parallel Chat API (primary) or Perplexity sonar-pro-search (academic paper searches). Automatically routes queries to the best backend. Use for finding papers, gathering research data, and verifying scientific information.","https:\u002F\u002Fgithub.com\u002Feric861129\u002FSKILLS_All-in-one\u002Fblob\u002Fmain\u002Fpublic\u002FSKILLS\u002FWriting & Research\u002Fresearch-lookup\u002FSKILL.md",[222,225,226,227],{"categoryNo":223,"categoryName":224,"taxonomyType":21,"taxonomyTypeDesc":22},"01JBSEEDCAT000000000000010","后端与接口",{"categoryNo":125,"categoryName":126,"taxonomyType":21,"taxonomyTypeDesc":22},{"categoryNo":19,"categoryName":20,"taxonomyType":21,"taxonomyTypeDesc":22},{"categoryNo":24,"categoryName":25,"taxonomyType":26,"taxonomyTypeDesc":27},[134,229,230,231,207,232,233,140,41,234,108,109,142,189,235,236,237,238,239,144],"lookup","research lookup","search","perplexity","api","openrouter","backend","up","look","queries","chat"]