[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"skill:01M6MCN7YVTKK8E2P63EPVEYQ7":3,"skill-same:01M6MCN7YVTKK8E2P63EPVEYQ7":82},{"skillNo":4,"name":5,"nameZh":6,"description":7,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":6,"repositoryNo":12,"starsCount":13,"licenseSpdxId":14,"licenseFileUrl":6,"sourceHtmlUrl":15,"categories":16,"keywords":22,"publishedAt":43,"linkOnly":44,"markdownContent":45,"markdownBody":46,"frontMatter":47,"refName":62,"commitSha":63,"skillDirectory":64,"forksCount":13,"repositoryHtmlUrl":65,"repositoryHomepageUrl":6,"ownerHtmlUrl":66,"ownerBlogUrl":6,"ownerDescription":6,"updatedAt":67,"install":68,"resources":81},"01M6MCN7YVTKK8E2P63EPVEYQ7","llm-wiki",null,"Use when building or maintaining a persistent personal knowledge base (second brain) in Obsidian where an LLM incrementally ingests sources, updates entity\u002Fconcept pages, maintains cross-references, and keeps a synthesis current. Triggers include \"second brain\", \"Obsidian wiki\", \"personal knowledge management\", \"ingest this paper\u002Farticle\u002Fbook\", \"build a research wiki\", \"compound knowledge\", \"Memex\", or whenever the user wants knowledge to accumulate across sessions instead of being re-derived by RAG on every query.","JiayiV\u002Fquestion_mixing","JiayiV","https:\u002F\u002Favatars.githubusercontent.com\u002Fu\u002F54466889?v=4","01M6JFJYYMRZ3QRGP9339JNQ2K","01M6JFJZ35KB0QVT50NAHPJA7G",0,"MIT","https:\u002F\u002Fgithub.com\u002FJiayiV\u002Fquestion_mixing\u002Fblob\u002Fmain\u002Fskills\u002Fpublic-02-claude-skills-main-engineering-llm-wiki-skills-llm-wiki-skill-e96f855f\u002FSKILL.md",[17],{"categoryNo":18,"categoryName":19,"taxonomyType":20,"taxonomyTypeDesc":21},"01JBSEEDCAT000000000000021","资料检索与研究","SCENARIO","使用场景",[23,24,25,26,27,28,29,30,31,32,33,34,35,36,37,38,39,40,41,42],"wiki","llm wiki","opencode","updates","re","sources","article","brain","across","knowledge base","query","references","cli","knowledge management","karpathy","knowledge-management","knowledge","code","rag","obsidian","2026-11-10 10:12:30",false,"---\nname: llm-wiki\ndescription: Use when building or maintaining a persistent personal knowledge base (second brain) in Obsidian where an LLM incrementally ingests sources, updates entity\u002Fconcept pages, maintains cross-references, and keeps a synthesis current. Triggers include \"second brain\", \"Obsidian wiki\", \"personal knowledge management\", \"ingest this paper\u002Farticle\u002Fbook\", \"build a research wiki\", \"compound knowledge\", \"Memex\", or whenever the user wants knowledge to accumulate across sessions instead of being re-derived by RAG on every query.\ncontext: fork\nversion: 2.9.0\nauthor: claude-code-skills\nlicense: MIT\ntags: [knowledge-management, obsidian, second-brain, pkm, rag-alternative, wiki, karpathy, memex]\ncompatible_tools: [claude-code, codex-cli, cursor, antigravity, opencode, gemini-cli]\n---\n\n# LLM Wiki — Second Brain for Claude Code + Obsidian\n\nInspired by Andrej Karpathy's LLM Wiki pattern ([gist](https:\u002F\u002Fgist.github.com\u002Fkarpathy\u002F442a6bf555914893e9891c11519de94f)). This skill turns Claude Code (or any agent CLI) into a disciplined wiki maintainer that **incrementally builds and maintains** a persistent, interlinked Obsidian vault as you feed it sources. The knowledge compounds — cross-references, contradictions, and synthesis are already there when you query.\n\n## Core principle\n\nMost LLM+docs workflows are **RAG**: retrieve fragments at query time, synthesize from scratch, forget. The wiki is **compounding**: sources are read once, integrated into a persistent markdown knowledge base, and kept current. You curate and ask; the LLM reads, files, cross-references, and maintains.\n\n> Obsidian is the IDE. The LLM is the programmer. The wiki is the codebase.\n\n## When to use\n\n- **Personal**: track goals, health, psychology, journaling, self-improvement\n- **Research**: deep dives over weeks on a topic — papers, articles, reports, evolving thesis\n- **Book companion**: file chapters as you read; build a fan-wiki-style companion for characters, themes, plot threads\n- **Business\u002Fteam**: internal wiki fed by Slack, meeting notes, calls — LLM does maintenance nobody else wants to do\n- **Competitive analysis, due diligence, trip planning, course notes, hobby deep-dives**\n\n**Do NOT use when:** you need one-shot Q&A over a fixed document (use RAG), you don't plan to add sources over time, or you don't want Obsidian in the loop.\n\n## Architecture (three layers)\n\n```\nvault\u002F\n├── raw\u002F                    # Layer 1 — IMMUTABLE source of truth\n│   ├── \u003Csource files>      # Articles, papers, PDFs, images, data\n│   └── assets\u002F             # Downloaded images from clipped articles\n├── wiki\u002F                   # Layer 2 — LLM-owned knowledge base\n│   ├── index.md            # Content catalog (LLM updates every ingest)\n│   ├── log.md              # Append-only timeline (## [YYYY-MM-DD] \u003Cop> | \u003Ctitle>)\n│   ├── entities\u002F           # Person\u002FOrg\u002FPlace pages\n│   ├── concepts\u002F           # Ideas, theories, frameworks\n│   ├── sources\u002F            # One summary page per ingested source\n│   ├── comparisons\u002F        # Cross-source analysis pages\n│   └── synthesis\u002F          # High-level syntheses, theses, overviews\n├── CLAUDE.md               # Schema + conventions (Claude Code)\n└── AGENTS.md               # Same content, for Codex\u002FCursor\u002FAntigravity\n```\n\n- **Layer 1 (raw\u002F)** — you own. LLM only reads; never writes.\n- **Layer 2 (wiki\u002F)** — LLM owns. It creates, updates, and cross-references pages. You read it.\n- **Layer 3 (CLAUDE.md \u002F AGENTS.md)** — the *schema*. Conventions, workflows, frontmatter rules. Co-evolved by you and the LLM.\n\n## Three core operations\n\n1. **Ingest** — LLM reads a source, discusses takeaways with you, writes a source summary, updates 10-15 relevant pages, updates index, appends to log. See `references\u002Fingest-workflow.md`.\n2. **Query** — LLM reads `index.md` first, drills into relevant pages, synthesizes with citations. Good answers get **filed back into the wiki** so explorations compound. See `references\u002Fquery-workflow.md`.\n3. **Lint** — Health check: contradictions, stale claims, orphan pages, missing cross-refs, concepts mentioned but lacking their own page, data gaps to fill with web search. See `references\u002Flint-workflow.md`.\n\n## Quick start\n\n```bash\n# 1. Initialize a vault (in Obsidian's vault directory)\npython scripts\u002Finit_vault.py --path ~\u002Fvaults\u002Fresearch --topic \"LLM interpretability\"\n\n# 2. Drop a source into raw\u002F, then ingest\n\u002Fwiki-ingest ~\u002Fvaults\u002Fresearch\u002Fraw\u002Fanthropic-monosemanticity.pdf\n\n# 3. Ask questions (answers can be re-filed into the wiki)\n\u002Fwiki-query \"how does monosemanticity compare to mechanistic interpretability?\"\n\n# 4. Periodic health check\n\u002Fwiki-lint\n\n# 5. See the timeline\n\u002Fwiki-log --last 10\n```\n\n## Slash commands (this plugin ships)\n\n| Command | Purpose |\n|---|---|\n| `\u002Fwiki-init` | Bootstrap a fresh vault with schema files + starter structure |\n| `\u002Fwiki-ingest \u003Cpath>` | Read a source, discuss, update wiki, log it |\n| `\u002Fwiki-query \u003Cquestion>` | Search wiki, synthesize answer, offer to file back |\n| `\u002Fwiki-lint` | Run health check — contradictions, orphans, stale claims, gaps |\n| `\u002Fwiki-log` | Show recent log entries (uses unix tools on `log.md`) |\n\n## Sub-agents (this plugin ships)\n\n| Agent | When dispatched |\n|---|---|\n| `wiki-ingestor` | Delegated ingest flow — reads source, proposes updates, applies after your approval |\n| `wiki-linter` | Runs the health-check workflow independently, reports findings |\n| `wiki-librarian` | Answers queries using index-first search, synthesizes with citations |\n\n## Python tools (`scripts\u002F`)\n\nAll tools are **standard library only** (no pip installs). Run with `python scripts\u002F\u003Ctool>.py --help`.\n\n| Script | Purpose |\n|---|---|\n| `init_vault.py` | Create folder structure + seed CLAUDE.md, AGENTS.md, index.md, log.md |\n| `ingest_source.py` | Helper: extract text\u002Ffrontmatter from a source file, ready for LLM review |\n| `update_index.py` | Regenerate `index.md` from wiki page frontmatter (category, date, source count) |\n| `append_log.py` | Append a standardized log entry `## [YYYY-MM-DD] \u003Cop> \\| \u003Ctitle>` |\n| `wiki_search.py` | BM25 search over wiki pages (standalone fallback when index.md isn't enough) |\n| `lint_wiki.py` | Find orphans (no inbound links), stale pages, missing cross-refs, broken links |\n| `graph_analyzer.py` | Compute link graph stats — hubs, orphans, clusters, disconnected components |\n| `export_marp.py` | Render a wiki page (or subtree) to a Marp slide deck |\n\n## Cross-tool compatibility\n\nThe vault's **schema** lives in CLAUDE.md (Claude Code) or AGENTS.md (Codex\u002FCursor\u002FAntigravity\u002FOpenCode). The same content works in both. This plugin ships both templates. For per-tool setup instructions see `references\u002Fcross-tool-setup.md`.\n\n```\nCLAUDE.md       → Claude Code\nAGENTS.md       → Codex CLI, Cursor, Antigravity, OpenCode, Gemini CLI\n.cursorrules    → legacy Cursor (pre-AGENTS.md)\n```\n\nThe scripts are pure Python stdlib → run identically everywhere. Only the loader file changes per tool.\n\n## Obsidian setup (recommended)\n\n- **Obsidian Web Clipper** — browser extension; converts web articles to markdown and drops them in `raw\u002F`\n- **Download images locally** — Settings → Files and links → Attachment folder path = `raw\u002Fassets\u002F`. Settings → Hotkeys → bind \"Download attachments for current file\" to `Ctrl+Shift+D`\n- **Graph view** — see hubs\u002Forphans; essential for spotting structural problems\n- **Marp plugin** — Markdown-based slide decks directly from wiki pages\n- **Dataview plugin** — dynamic tables\u002Flists over page frontmatter (tags, dates, source counts)\n- **Git** — the vault is a plain markdown repo; version it\n\nFull setup walkthrough: `references\u002Fobsidian-setup.md`\n\n## Why this works (vs plain RAG)\n\n| Plain RAG | LLM Wiki |\n|---|---|\n| Rediscover knowledge each query | Knowledge accumulates |\n| Cross-references re-computed every time | Cross-references pre-written and maintained |\n| Contradictions surface only if you ask | Contradictions flagged during ingest |\n| Exploration disappears into chat history | Good answers re-filed as new pages |\n| Scales by embeddings infrastructure | Scales by markdown + `index.md` + optional local search |\n\nAt ~100 sources \u002F hundreds of pages, `index.md` + filesystem search is enough. Past that, layer in a local search tool like [qmd](https:\u002F\u002Fgithub.com\u002Ftobi\u002Fqmd) or use `scripts\u002Fwiki_search.py`.\n\n## Related skills (chains via `context: fork`)\n\nThis skill is marked `context: fork` so other skills can chain into it:\n\n- **`para-memory-files`** — PARA-method memory; complementary as long-term personal memory that feeds sources into the wiki\n- **`obsidian-vault`** (mattpocock) — lightweight Obsidian note helper; this skill is the maintained-wiki layer on top\n- **`rag-design`** — when wiki outgrows ~500 pages, use rag-design to bolt on a retrieval layer\n- **`mcp-design`** — expose the wiki as an MCP tool\n- **`agent-communication`** — for multi-agent wiki maintenance (ingestor + linter + librarian)\n\n## Reference docs\n\n- `references\u002Fwiki-schema.md` — full vault layout, page frontmatter, naming conventions\n- `references\u002Fpage-formats.md` — entity, concept, source, comparison, synthesis templates\n- `references\u002Fingest-workflow.md` — the detailed ingest flow the wiki-ingestor agent follows\n- `references\u002Fquery-workflow.md` — query patterns, citation format, re-filing answers\n- `references\u002Flint-workflow.md` — health-check heuristics\n- `references\u002Fobsidian-setup.md` — Obsidian plugins, hotkeys, vault config\n- `references\u002Fcross-tool-setup.md` — per-tool setup (Codex, Cursor, Antigravity, etc.)\n- `references\u002Fmemex-principles.md` — Bush's Memex, why the LLM changes the maintenance math\n\n## Templates (`assets\u002F`)\n\n- `CLAUDE.md.template`, `AGENTS.md.template`, `.cursorrules.template` — schema loaders per tool\n- `index.md.template`, `log.md.template` — starter index and log\n- `page-templates\u002F` — entity, concept, source-summary, comparison, synthesis\n- `example-vault\u002F` — small worked example you can study or copy\n\n## Iron rule\n\n**The LLM never edits files in `raw\u002F`.** Ever. Sources are immutable. All LLM writes go to `wiki\u002F`. If you need to correct a source, do it in `raw\u002F` yourself — then re-ingest.\n","# LLM Wiki — Second Brain for Claude Code + Obsidian\n\nInspired by Andrej Karpathy's LLM Wiki pattern ([gist](https:\u002F\u002Fgist.github.com\u002Fkarpathy\u002F442a6bf555914893e9891c11519de94f)). This skill turns Claude Code (or any agent CLI) into a disciplined wiki maintainer that **incrementally builds and maintains** a persistent, interlinked Obsidian vault as you feed it sources. The knowledge compounds — cross-references, contradictions, and synthesis are already there when you query.\n\n## Core principle\n\nMost LLM+docs workflows are **RAG**: retrieve fragments at query time, synthesize from scratch, forget. The wiki is **compounding**: sources are read once, integrated into a persistent markdown knowledge base, and kept current. You curate and ask; the LLM reads, files, cross-references, and maintains.\n\n> Obsidian is the IDE. The LLM is the programmer. The wiki is the codebase.\n\n## When to use\n\n- **Personal**: track goals, health, psychology, journaling, self-improvement\n- **Research**: deep dives over weeks on a topic — papers, articles, reports, evolving thesis\n- **Book companion**: file chapters as you read; build a fan-wiki-style companion for characters, themes, plot threads\n- **Business\u002Fteam**: internal wiki fed by Slack, meeting notes, calls — LLM does maintenance nobody else wants to do\n- **Competitive analysis, due diligence, trip planning, course notes, hobby deep-dives**\n\n**Do NOT use when:** you need one-shot Q&A over a fixed document (use RAG), you don't plan to add sources over time, or you don't want Obsidian in the loop.\n\n## Architecture (three layers)\n\n```\nvault\u002F\n├── raw\u002F                    # Layer 1 — IMMUTABLE source of truth\n│   ├── \u003Csource files>      # Articles, papers, PDFs, images, data\n│   └── assets\u002F             # Downloaded images from clipped articles\n├── wiki\u002F                   # Layer 2 — LLM-owned knowledge base\n│   ├── index.md            # Content catalog (LLM updates every ingest)\n│   ├── log.md              # Append-only timeline (## [YYYY-MM-DD] \u003Cop> | \u003Ctitle>)\n│   ├── entities\u002F           # Person\u002FOrg\u002FPlace pages\n│   ├── concepts\u002F           # Ideas, theories, frameworks\n│   ├── sources\u002F            # One summary page per ingested source\n│   ├── comparisons\u002F        # Cross-source analysis pages\n│   └── synthesis\u002F          # High-level syntheses, theses, overviews\n├── CLAUDE.md               # Schema + conventions (Claude Code)\n└── AGENTS.md               # Same content, for Codex\u002FCursor\u002FAntigravity\n```\n\n- **Layer 1 (raw\u002F)** — you own. LLM only reads; never writes.\n- **Layer 2 (wiki\u002F)** — LLM owns. It creates, updates, and cross-references pages. You read it.\n- **Layer 3 (CLAUDE.md \u002F AGENTS.md)** — the *schema*. Conventions, workflows, frontmatter rules. Co-evolved by you and the LLM.\n\n## Three core operations\n\n1. **Ingest** — LLM reads a source, discusses takeaways with you, writes a source summary, updates 10-15 relevant pages, updates index, appends to log. See `references\u002Fingest-workflow.md`.\n2. **Query** — LLM reads `index.md` first, drills into relevant pages, synthesizes with citations. Good answers get **filed back into the wiki** so explorations compound. See `references\u002Fquery-workflow.md`.\n3. **Lint** — Health check: contradictions, stale claims, orphan pages, missing cross-refs, concepts mentioned but lacking their own page, data gaps to fill with web search. See `references\u002Flint-workflow.md`.\n\n## Quick start\n\n```bash\n# 1. Initialize a vault (in Obsidian's vault directory)\npython scripts\u002Finit_vault.py --path ~\u002Fvaults\u002Fresearch --topic \"LLM interpretability\"\n\n# 2. Drop a source into raw\u002F, then ingest\n\u002Fwiki-ingest ~\u002Fvaults\u002Fresearch\u002Fraw\u002Fanthropic-monosemanticity.pdf\n\n# 3. Ask questions (answers can be re-filed into the wiki)\n\u002Fwiki-query \"how does monosemanticity compare to mechanistic interpretability?\"\n\n# 4. Periodic health check\n\u002Fwiki-lint\n\n# 5. See the timeline\n\u002Fwiki-log --last 10\n```\n\n## Slash commands (this plugin ships)\n\n| Command | Purpose |\n|---|---|\n| `\u002Fwiki-init` | Bootstrap a fresh vault with schema files + starter structure |\n| `\u002Fwiki-ingest \u003Cpath>` | Read a source, discuss, update wiki, log it |\n| `\u002Fwiki-query \u003Cquestion>` | Search wiki, synthesize answer, offer to file back |\n| `\u002Fwiki-lint` | Run health check — contradictions, orphans, stale claims, gaps |\n| `\u002Fwiki-log` | Show recent log entries (uses unix tools on `log.md`) |\n\n## Sub-agents (this plugin ships)\n\n| Agent | When dispatched |\n|---|---|\n| `wiki-ingestor` | Delegated ingest flow — reads source, proposes updates, applies after your approval |\n| `wiki-linter` | Runs the health-check workflow independently, reports findings |\n| `wiki-librarian` | Answers queries using index-first search, synthesizes with citations |\n\n## Python tools (`scripts\u002F`)\n\nAll tools are **standard library only** (no pip installs). Run with `python scripts\u002F\u003Ctool>.py --help`.\n\n| Script | Purpose |\n|---|---|\n| `init_vault.py` | Create folder structure + seed CLAUDE.md, AGENTS.md, index.md, log.md |\n| `ingest_source.py` | Helper: extract text\u002Ffrontmatter from a source file, ready for LLM review |\n| `update_index.py` | Regenerate `index.md` from wiki page frontmatter (category, date, source count) |\n| `append_log.py` | Append a standardized log entry `## [YYYY-MM-DD] \u003Cop> \\| \u003Ctitle>` |\n| `wiki_search.py` | BM25 search over wiki pages (standalone fallback when index.md isn't enough) |\n| `lint_wiki.py` | Find orphans (no inbound links), stale pages, missing cross-refs, broken links |\n| `graph_analyzer.py` | Compute link graph stats — hubs, orphans, clusters, disconnected components |\n| `export_marp.py` | Render a wiki page (or subtree) to a Marp slide deck |\n\n## Cross-tool compatibility\n\nThe vault's **schema** lives in CLAUDE.md (Claude Code) or AGENTS.md (Codex\u002FCursor\u002FAntigravity\u002FOpenCode). The same content works in both. This plugin ships both templates. For per-tool setup instructions see `references\u002Fcross-tool-setup.md`.\n\n```\nCLAUDE.md       → Claude Code\nAGENTS.md       → Codex CLI, Cursor, Antigravity, OpenCode, Gemini CLI\n.cursorrules    → legacy Cursor (pre-AGENTS.md)\n```\n\nThe scripts are pure Python stdlib → run identically everywhere. Only the loader file changes per tool.\n\n## Obsidian setup (recommended)\n\n- **Obsidian Web Clipper** — browser extension; converts web articles to markdown and drops them in `raw\u002F`\n- **Download images locally** — Settings → Files and links → Attachment folder path = `raw\u002Fassets\u002F`. Settings → Hotkeys → bind \"Download attachments for current file\" to `Ctrl+Shift+D`\n- **Graph view** — see hubs\u002Forphans; essential for spotting structural problems\n- **Marp plugin** — Markdown-based slide decks directly from wiki pages\n- **Dataview plugin** — dynamic tables\u002Flists over page frontmatter (tags, dates, source counts)\n- **Git** — the vault is a plain markdown repo; version it\n\nFull setup walkthrough: `references\u002Fobsidian-setup.md`\n\n## Why this works (vs plain RAG)\n\n| Plain RAG | LLM Wiki |\n|---|---|\n| Rediscover knowledge each query | Knowledge accumulates |\n| Cross-references re-computed every time | Cross-references pre-written and maintained |\n| Contradictions surface only if you ask | Contradictions flagged during ingest |\n| Exploration disappears into chat history | Good answers re-filed as new pages |\n| Scales by embeddings infrastructure | Scales by markdown + `index.md` + optional local search |\n\nAt ~100 sources \u002F hundreds of pages, `index.md` + filesystem search is enough. Past that, layer in a local search tool like [qmd](https:\u002F\u002Fgithub.com\u002Ftobi\u002Fqmd) or use `scripts\u002Fwiki_search.py`.\n\n## Related skills (chains via `context: fork`)\n\nThis skill is marked `context: fork` so other skills can chain into it:\n\n- **`para-memory-files`** — PARA-method memory; complementary as long-term personal memory that feeds sources into the wiki\n- **`obsidian-vault`** (mattpocock) — lightweight Obsidian note helper; this skill is the maintained-wiki layer on top\n- **`rag-design`** — when wiki outgrows ~500 pages, use rag-design to bolt on a retrieval layer\n- **`mcp-design`** — expose the wiki as an MCP tool\n- **`agent-communication`** — for multi-agent wiki maintenance (ingestor + linter + librarian)\n\n## Reference docs\n\n- `references\u002Fwiki-schema.md` — full vault layout, page frontmatter, naming conventions\n- `references\u002Fpage-formats.md` — entity, concept, source, comparison, synthesis templates\n- `references\u002Fingest-workflow.md` — the detailed ingest flow the wiki-ingestor agent follows\n- `references\u002Fquery-workflow.md` — query patterns, citation format, re-filing answers\n- `references\u002Flint-workflow.md` — health-check heuristics\n- `references\u002Fobsidian-setup.md` — Obsidian plugins, hotkeys, vault config\n- `references\u002Fcross-tool-setup.md` — per-tool setup (Codex, Cursor, Antigravity, etc.)\n- `references\u002Fmemex-principles.md` — Bush's Memex, why the LLM changes the maintenance math\n\n## Templates (`assets\u002F`)\n\n- `CLAUDE.md.template`, `AGENTS.md.template`, `.cursorrules.template` — schema loaders per tool\n- `index.md.template`, `log.md.template` — starter index and log\n- `page-templates\u002F` — entity, concept, source-summary, comparison, synthesis\n- `example-vault\u002F` — small worked example you can study or copy\n\n## Iron rule\n\n**The LLM never edits files in `raw\u002F`.** Ever. Sources are immutable. All LLM writes go to `wiki\u002F`. If you need to correct a source, do it in `raw\u002F` yourself — then re-ingest.\n",{"name":5,"tags":48,"author":53,"context":54,"license":14,"version":55,"description":7,"compatible_tools":56},[38,42,49,50,51,23,37,52],"second-brain","pkm","rag-alternative","memex","claude-code-skills","fork","2.9.0",[57,58,59,60,25,61],"claude-code","codex-cli","cursor","antigravity","gemini-cli","main","068c2486ccc0f8d200a3db4d21ec137064eda3b4","skills\u002Fpublic-02-claude-skills-main-engineering-llm-wiki-skills-llm-wiki-skill-e96f855f","https:\u002F\u002Fgithub.com\u002FJiayiV\u002Fquestion_mixing","https:\u002F\u002Fgithub.com\u002FJiayiV","2026-08-25 16:30:52",{"source":8,"skillName":5,"commands":69,"prompt":79,"zipAvailable":80},[70,73,76],{"tool":71,"command":72},"npx","npx skills add JiayiV\u002Fquestion_mixing --skill llm-wiki",{"tool":74,"command":75},"bunx","bunx skills add JiayiV\u002Fquestion_mixing --skill llm-wiki",{"tool":77,"command":78},"pnpm","pnpm dlx skills add JiayiV\u002Fquestion_mixing --skill llm-wiki","安装技能，技能地址：https:\u002F\u002Fgithub.com\u002FJiayiV\u002Fquestion_mixing\u002Fblob\u002Fmain\u002Fskills\u002Fpublic-02-claude-skills-main-engineering-llm-wiki-skills-llm-wiki-skill-e96f855f\u002FSKILL.md",true,[],[83,108,135,166,194,229],{"skillNo":84,"name":85,"nameZh":6,"description":86,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":6,"repositoryNo":12,"starsCount":13,"licenseSpdxId":14,"licenseFileUrl":6,"sourceHtmlUrl":87,"categories":88,"keywords":89,"publishedAt":43},"01M6NN9PJS9NHWPDG4QP93BP8F","webmcpify","Make a web app agent-ready — propose a WebMCP tool manifest, integrate, verify in a real browser, heal; unrelated code stays untouched. Use for \"webmcpify\", \"add WebMCP\", or \"expose app actions to AI agents\".","https:\u002F\u002Fgithub.com\u002FJiayiV\u002Fquestion_mixing\u002Fblob\u002Fmain\u002Fskills\u002Fwebmcpify\u002FSKILL.md",[],[85,90,91,92,93,40,94,95,96,97,98,99,100,101,102,103,104,105,106,107],"github","ai agents","web","notes","browser","https","actions","manifest","inventory","verify","tool","use","ready","unrelated","app","scope","for","webmcp",{"skillNo":109,"name":110,"nameZh":6,"description":111,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":6,"repositoryNo":12,"starsCount":13,"licenseSpdxId":112,"licenseFileUrl":6,"sourceHtmlUrl":113,"categories":114,"keywords":115,"publishedAt":43},"01M6NN6KZE5RBN6NGHN2WXV1RE","vardoger-analyze","Use when the user asks to personalize the GitHub Copilot CLI assistant, adapt Copilot to their style, use vardoger, or analyze their Copilot CLI conversation history. Reads the local session directory at `~\u002F.copilot\u002Fsession-state\u002F`, extracts recurring preferences and conventions, and writes a fenced personalization block into `~\u002F.copilot\u002Fcopilot-instructions.md`. Runs entirely on the user's machine via the local `vardoger` CLI (`pipx install vardoger`); no network calls and no uploads. Triggers: 'personalize my copilot', 'analyze my copilot history', 'tailor copilot to me', 'run vardoger', 'update my copilot instructions from history', 'make copilot learn my style'.","Apache-2.0","https:\u002F\u002Fgithub.com\u002FJiayiV\u002Fquestion_mixing\u002Fblob\u002Fmain\u002Fskills\u002Fvardoger-analyze\u002FSKILL.md",[],[116,117,118,35,119,120,121,122,123,124,125,126,127,128,129,130,131,132,133,134],"vardoger analyze","analyze","directory","github copilot","copilot","run","update","learn","network","my style","install","conversation","me","local","github copilot cli","pipx","my","at","copilot instructions",{"skillNo":136,"name":137,"nameZh":6,"description":138,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":6,"repositoryNo":12,"starsCount":13,"licenseSpdxId":14,"licenseFileUrl":6,"sourceHtmlUrl":139,"categories":140,"keywords":147,"publishedAt":43},"01M6NN3CZGQP3XAJ04BM1NNYGT","transloadit-media-processing","Process media files (video, audio, images, documents) using Transloadit. Use when asked to encode video to HLS\u002FMP4, generate thumbnails, resize or watermark images, extract audio, concatenate clips, add subtitles, OCR documents, or run any media processing pipeline. Covers 86+ processing robots for file transformation at scale.","https:\u002F\u002Fgithub.com\u002FJiayiV\u002Fquestion_mixing\u002Fblob\u002Fmain\u002Fskills\u002Ftransloadit-media-processing\u002FSKILL.md",[141,144],{"categoryNo":142,"categoryName":143,"taxonomyType":20,"taxonomyTypeDesc":21},"01JBSEEDCAT000000000000016","图像与多媒体",{"categoryNo":145,"categoryName":146,"taxonomyType":20,"taxonomyTypeDesc":21},"01JBSEEDCAT000000000000023","Skill与智能体开发",[148,149,150,151,152,153,154,155,156,157,158,159,160,101,161,162,163,164,106,165],"transloadit media processing","Media","media processing","processing","MCP","mcp server","documents","subtitles","pipeline","ocr","Video","Audio","mp4","images","files","transformation","free","hls",{"skillNo":167,"name":168,"nameZh":6,"description":169,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":6,"repositoryNo":12,"starsCount":13,"licenseSpdxId":14,"licenseFileUrl":6,"sourceHtmlUrl":170,"categories":171,"keywords":175,"publishedAt":43},"01M6NN28E92F9GH4WFYQ2H1Y15","terraform-azurerm-set-diff-analyzer","Analyze Terraform plan JSON output for AzureRM Provider to distinguish between false-positive diffs (order-only changes in Set-type attributes) and actual resource changes. Use when reviewing terraform plan output for Azure resources like Application Gateway, Load Balancer, Firewall, Front Door, NSG, and other resources with Set-type attributes that cause spurious diffs due to internal ordering changes.","https:\u002F\u002Fgithub.com\u002FJiayiV\u002Fquestion_mixing\u002Fblob\u002Fmain\u002Fskills\u002Fterraform-azurerm-set-diff-analyzer\u002FSKILL.md",[172],{"categoryNo":173,"categoryName":174,"taxonomyType":20,"taxonomyTypeDesc":21},"01JBSEEDCAT000000000000011","运维与部署",[176,177,178,179,180,181,182,183,184,101,185,186,187,188,189,190,191,192,193,106],"terraform","SET","terraform azurerm set diff analyzer","diff","analyzer","diffs","azure","gateway","firewall","only","resource changes","provider","resources","load balancer","attributes","front","load","balancer",{"skillNo":195,"name":196,"nameZh":6,"description":197,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":6,"repositoryNo":12,"starsCount":13,"licenseSpdxId":14,"licenseFileUrl":6,"sourceHtmlUrl":198,"categories":199,"keywords":211,"publishedAt":43},"01M6NN1QMH2TDYWTH3XY3N5XPV","technical-job-search","Use this skill when a software engineer asks for help with job search tasks: parsing or analyzing a job description, tailoring a CV\u002Fresume, writing a cover letter, evaluating a job offer, or drafting a post-interview follow-up email. Do not activate for general career advice unrelated to an active job search action.","https:\u002F\u002Fgithub.com\u002FJiayiV\u002Fquestion_mixing\u002Fblob\u002Fmain\u002Fskills\u002Ftechnical-job-search\u002FSKILL.md",[200,203,208],{"categoryNo":201,"categoryName":202,"taxonomyType":20,"taxonomyTypeDesc":21},"01JBSEEDCAT000000000000018","写作与文案",{"categoryNo":204,"categoryName":205,"taxonomyType":206,"taxonomyTypeDesc":207},"01JBSEEDCAT000000000000036","内容创作者","PROFESSION","适用职业",{"categoryNo":209,"categoryName":210,"taxonomyType":206,"taxonomyTypeDesc":207},"01JBSEEDCAT000000000000038","人力资源",[212,213,214,215,216,217,218,219,117,220,101,221,222,223,224,225,226,227,106,228],"search","Job Search","technical job search","technical","writing","tasks","email","parsing","interview","up","engineer","general","activate","do","task","cv","letter",{"skillNo":230,"name":231,"nameZh":6,"description":232,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":6,"repositoryNo":12,"starsCount":13,"licenseSpdxId":14,"licenseFileUrl":6,"sourceHtmlUrl":233,"categories":234,"keywords":235,"publishedAt":43},"01M6NMZFA67XYS1GS09WT42235","steno-mode","Shorthand-first response compression that cuts ~40% of response tokens while preserving technical precision and exact literals. Use when the user says \"steno mode\", \"shorthand mode\", \"compressed responses\", \"token reduction\", \"brief structured output\", or invokes \u002Fsteno. Supports four compression levels: lite, brief, court, machine. Do not trigger for requests needing polished prose such as onboarding\u002Ftutorial content, stakeholder or customer-facing copy, or teaching-focused explanations.","https:\u002F\u002Fgithub.com\u002FJiayiV\u002Fquestion_mixing\u002Fblob\u002Fmain\u002Fskills\u002Fsteno-mode\u002FSKILL.md",[],[236,237,238,239,240,215,241,101,225,242,243,244,245,246,247,248,249,250,106,251],"steno mode","mode","customer","brief","first","tutorial","teaching","tokens","as","trigger","structured output","machine","output","court","requests","response compression"]