[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"skill:01M6ADYF8W0422TH7R3A4DD85B":3,"skill-same:01M6ADYF8W0422TH7R3A4DD85B":103},{"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":26,"publishedAt":55,"linkOnly":56,"markdownContent":57,"markdownBody":58,"frontMatter":59,"refName":62,"commitSha":63,"skillDirectory":64,"forksCount":65,"repositoryHtmlUrl":66,"repositoryHomepageUrl":67,"ownerHtmlUrl":68,"ownerBlogUrl":69,"ownerDescription":70,"updatedAt":71,"install":72,"resources":85},"01M6ADYF8W0422TH7R3A4DD85B","pinecone-rag",null,"Build production RAG pipelines and persistent agent memory using Pinecone as the vector database backend. ALWAYS USE THIS SKILL when the user mentions Pinecone, wants to index documents for semantic search, build a retrieval-augmented generation system, store agent memory across sessions, implement hybrid search, or connect an LLM to a searchable knowledge base — even if they don't say \"Pinecone\" explicitly. Also use when the user asks about vector databases for RAG, namespace isolation for multi-tenant agents, embedding pipelines, or scaling a knowledge base beyond what local storage can handle. DO NOT use for local-only vector stores (Chroma, FAISS, pgvector) or pure keyword search with no semantic component.","github\u002Fawesome-copilot","github","https:\u002F\u002Favatars.githubusercontent.com\u002Fu\u002F9919?v=4","01M6AC83DS133TPXFSP1DHNCC6","GitHub","01M6AD3PBDYPK7NVFJNJ66Y8RS",41046,"Apache-2.0","https:\u002F\u002Fgithub.com\u002Fgithub\u002Fawesome-copilot\u002Fblob\u002Fmain\u002Fskills\u002Fpinecone-rag\u002FSKILL.md",[18,23],{"categoryNo":19,"categoryName":20,"taxonomyType":21,"taxonomyTypeDesc":22},"01JBSEEDCAT000000000000010","后端与接口","SCENARIO","使用场景",{"categoryNo":24,"categoryName":25,"taxonomyType":21,"taxonomyTypeDesc":22},"01JBSEEDCAT000000000000023","Skill与智能体开发",[27,28,29,30,31,32,33,34,35,36,37,38,39,40,41,42,43,44,45,46,47,48,49,50,51,52,53,54],"rag","pinecone","pinecone rag","vector","when","tenant","Hybrid Search","embedding","scaling","local","across","databases","only","vector stores","knowledge","python","search","index","database","agent memory","connect","semantic search","documents","sessions","component","storage","implement","memory","2026-11-10 10:12:30",false,"---\nname: pinecone-rag\ndescription: >\n  Build production RAG pipelines and persistent agent memory using Pinecone as\n  the vector database backend. ALWAYS USE THIS SKILL when the user mentions\n  Pinecone, wants to index documents for semantic search, build a\n  retrieval-augmented generation system, store agent memory across sessions,\n  implement hybrid search, or connect an LLM to a searchable knowledge base —\n  even if they don't say \"Pinecone\" explicitly. Also use when the user asks\n  about vector databases for RAG, namespace isolation for multi-tenant agents,\n  embedding pipelines, or scaling a knowledge base beyond what local storage\n  can handle. DO NOT use for local-only vector stores (Chroma, FAISS, pgvector)\n  or pure keyword search with no semantic component.\nlicense: Apache-2.0\ncompatibility: \"pinecone>=6.0.0, Python 3.10+\"\n---\n\n# Pinecone RAG Skill\n\nThis skill guides you through building a production RAG pipeline or persistent\nagent memory system using Pinecone. Follow the workflow from start to finish —\ndon't skip steps or jump to code before understanding what the user actually\nneeds.\n\n## Before you start — ask one question\n\nBefore writing any code, identify which of these two use cases applies:\n\n**A — RAG over documents**: User wants to index a corpus (PDFs, docs, code,\nweb pages) and retrieve relevant chunks to ground LLM responses.\n\n**B — Agent memory**: User wants an agent to remember facts, decisions, or\ncontext across sessions or across multiple agents sharing a knowledge base.\n\nThe setup is similar but the namespace strategy and retrieval patterns differ.\nIf the user hasn't said, ask: *\"Is this for document retrieval, agent memory,\nor both?\"* Then follow the relevant workflow below.\n\n---\n\n## Step 1 — Choose your index configuration\n\nPick the index type before writing any code. Getting this wrong means\nre-creating the index later.\n\n**Serverless (recommended for most cases)**\n```python\nfrom pinecone import Pinecone, ServerlessSpec\n\npc = Pinecone(api_key=\"PINECONE_API_KEY\")\n\nif \"my-index\" not in pc.list_indexes().names():\n    pc.create_index(\n        name=\"my-index\",\n        dimension=1536,        # must match your embedding model exactly\n        metric=\"cosine\",\n        spec=ServerlessSpec(cloud=\"aws\", region=\"us-east-1\")\n    )\nindex = pc.Index(\"my-index\")\n```\n\n**Pod-based (for consistent high-throughput production)**\n```python\nfrom pinecone import PodSpec\n\npc.create_index(\n    name=\"my-index-prod\",\n    dimension=1536,\n    metric=\"cosine\",\n    spec=PodSpec(environment=\"us-east1-gcp\", pod_type=\"p1.x1\")\n)\n```\n\n**Dimension quick reference — match this exactly to your embedding model:**\n| Model | Dimension |\n|---|---|\n| `text-embedding-3-small` | 1536 |\n| `text-embedding-3-large` | 3072 |\n| `voyage-3` \u002F `voyage-multimodal-3` | 1024 |\n| `BAAI\u002Fbge-large-en-v1.5` | 1024 |\n| `intfloat\u002Fmultilingual-e5-large` (Arabic, Malay, Chinese) | 1024 |\n\n> **Checkpoint**: Index exists, dimension matches embedding model, `index.describe_index_stats()` returns without error.\n\n---\n\n## Step 2 — Embed and upsert documents\n\nAlways batch upserts — never upsert one vector at a time.\n\n```python\nfrom openai import OpenAI\n\nclient = OpenAI()\n\ndef embed(texts: list[str]) -> list[list[float]]:\n    res = client.embeddings.create(model=\"text-embedding-3-small\", input=texts)\n    return [r.embedding for r in res.data]\n\ndef upsert_docs(index, docs: list[dict], namespace: str = \"default\"):\n    \"\"\"docs = [{\"id\": \"...\", \"text\": \"...\", \"metadata\": {...}}]\"\"\"\n    BATCH = 100\n    for i in range(0, len(docs), BATCH):\n        batch = docs[i:i + BATCH]\n        vecs = [\n            {\n                \"id\": d[\"id\"],\n                \"values\": emb,\n                \"metadata\": {**d.get(\"metadata\", {}), \"text\": d[\"text\"]}\n            }\n            for d, emb in zip(batch, embed([d[\"text\"] for d in batch]))\n        ]\n        index.upsert(vectors=vecs, namespace=namespace)\n```\n\n**Always store the original text in metadata** — this avoids a second lookup\nat retrieval time.\n\n> **Checkpoint**: `index.describe_index_stats()` shows vector count > 0 in the\n> target namespace.\n\n---\n\n## Step 3 — Choose retrieval strategy\n\n### Dense (semantic) search — use for most cases\n```python\ndef search(index, query: str, top_k: int = 5, namespace: str = \"default\",\n           filter: dict = None) -> list[dict]:\n    [q_emb] = embed([query])\n    results = index.query(\n        vector=q_emb, top_k=top_k, namespace=namespace,\n        include_metadata=True, filter=filter\n    )\n    return [{\"text\": m.metadata[\"text\"], \"score\": m.score, \"id\": m.id}\n            for m in results.matches]\n```\n\n### Hybrid search (semantic + BM25 keyword) — use when corpus has exact terminology\nUse hybrid when the domain has precise terms that semantic search misses:\nlegal citations, medical codes, product SKUs, API method names.\n\n```python\nfrom pinecone_text.sparse import BM25Encoder\n\nbm25 = BM25Encoder().default()\nbm25.fit([d[\"text\"] for d in docs])  # fit once on your corpus\n\ndef hybrid_search(index, query: str, top_k: int = 5, alpha: float = 0.7):\n    \"\"\"alpha=1.0 is pure dense; alpha=0.0 is pure sparse.\"\"\"\n    dense = [v * alpha for v in embed([query])[0]]\n    sparse_raw = bm25.encode_queries(query)\n    sparse = {\n        \"indices\": sparse_raw[\"indices\"],\n        \"values\": [v * (1 - alpha) for v in sparse_raw[\"values\"]]\n    }\n    return index.query(vector=dense, sparse_vector=sparse,\n                       top_k=top_k, include_metadata=True).matches\n```\n\n### Metadata filtering — use to scope results before semantic ranking\n```python\n# Exact match\nresults = index.query(vector=emb, filter={\"source\": {\"$eq\": \"confluence\"}})\n\n# Combined filter\nresults = index.query(vector=emb, filter={\n    \"$and\": [\n        {\"category\": {\"$eq\": \"engineering\"}},\n        {\"language\": {\"$in\": [\"en\", \"ar\"]}}\n    ]\n})\n```\n\n> **Checkpoint**: A test query returns relevant results with scores > 0.7 for\n> clearly matching content.\n\n---\n\n## Step 4A — Full RAG pipeline (document use case)\n\n```python\ndef rag_answer(index, question: str, namespace: str = \"default\",\n               model: str = \"gpt-4o-mini\") -> str:\n    hits = search(index, question, top_k=5, namespace=namespace)\n    context = \"\\n\\n\".join(h[\"text\"] for h in hits)\n\n    return client.chat.completions.create(\n        model=model,\n        messages=[\n            {\n                \"role\": \"system\",\n                \"content\": (\n                    \"Answer using only the provided context. \"\n                    \"If the answer isn't in the context, say so.\\n\\n\"\n                    f\"Context:\\n{context}\"\n                )\n            },\n            {\"role\": \"user\", \"content\": question}\n        ]\n    ).choices[0].message.content\n```\n\n---\n\n## Step 4B — Agent memory (memory use case)\n\nUse namespaces to isolate each agent's or user's memories completely.\nNamespace per agent prevents memory bleed across users or sessions.\n\n```python\nimport time, hashlib\n\ndef remember(index, agent_id: str, content: str,\n             memory_type: str = \"fact\"):\n    \"\"\"Store a memory for an agent.\"\"\"\n    mem_id = hashlib.md5(\n        f\"{agent_id}{content}{time.time()}\".encode()\n    ).hexdigest()\n    [emb] = embed([content])\n    index.upsert(\n        vectors=[{\n            \"id\": mem_id,\n            \"values\": emb,\n            \"metadata\": {\n                \"text\": content,\n                \"type\": memory_type,\n                \"timestamp\": time.time(),\n                \"agent_id\": agent_id\n            }\n        }],\n        namespace=f\"agent_{agent_id}\"\n    )\n\ndef recall(index, agent_id: str, query: str,\n           top_k: int = 5) -> list[str]:\n    \"\"\"Recall relevant memories for an agent.\"\"\"\n    return [h[\"text\"] for h in\n            search(index, query, top_k=top_k,\n                   namespace=f\"agent_{agent_id}\")]\n\ndef forget(index, agent_id: str):\n    \"\"\"Wipe all memories for an agent (e.g., on user request).\"\"\"\n    index.delete(delete_all=True, namespace=f\"agent_{agent_id}\")\n```\n\n---\n\n## Step 5 — Wire it together and test end to end\n\nRun a quick smoke test before integrating into the larger system:\n\n```python\n# Smoke test\nupsert_docs(index, [\n    {\"id\": \"t1\", \"text\": \"Pinecone is a vector database for semantic search.\"},\n    {\"id\": \"t2\", \"text\": \"RAG combines retrieval with language model generation.\"},\n])\n\nhits = search(index, \"What is Pinecone?\")\nassert hits[0][\"score\"] > 0.7, f\"Expected high similarity, got {hits[0]['score']}\"\nprint(\"Smoke test passed:\", hits[0][\"text\"])\n```\n\n> **Checkpoint**: Smoke test passes. End-to-end: index → upsert → query →\n> LLM response works without errors.\n\n---\n\n## Common pitfalls — fix these before they become bugs\n\n- **Dimension mismatch**: always verify `len(embed([\"test\"])[0])` matches\n  the index dimension before your first upsert.\n- **Missing text in metadata**: if you don't store `\"text\"` in metadata,\n  you'll need a second lookup to get the actual content at query time.\n- **Single-vector upserts in a loop**: always batch in chunks of 100.\n- **No namespace strategy**: decide upfront — one namespace per user\u002Fagent\n  prevents cross-tenant data leaks that are hard to fix later.\n- **Fitting BM25 on a small corpus**: BM25 needs a representative corpus to\n  build good term frequencies. Fit on at least a few hundred documents.\n\n## When NOT to use this skill\n\nUse a different approach when:\n- The dataset fits in memory and latency doesn't matter → use FAISS or Chroma\n- You're already on PostgreSQL and want to avoid a new service → use pgvector\n- You need sub-5ms p99 latency with no external API calls → local vector store\n- The user explicitly wants a different vector DB (Weaviate, Qdrant, etc.)","# Pinecone RAG Skill\n\nThis skill guides you through building a production RAG pipeline or persistent\nagent memory system using Pinecone. Follow the workflow from start to finish —\ndon't skip steps or jump to code before understanding what the user actually\nneeds.\n\n## Before you start — ask one question\n\nBefore writing any code, identify which of these two use cases applies:\n\n**A — RAG over documents**: User wants to index a corpus (PDFs, docs, code,\nweb pages) and retrieve relevant chunks to ground LLM responses.\n\n**B — Agent memory**: User wants an agent to remember facts, decisions, or\ncontext across sessions or across multiple agents sharing a knowledge base.\n\nThe setup is similar but the namespace strategy and retrieval patterns differ.\nIf the user hasn't said, ask: *\"Is this for document retrieval, agent memory,\nor both?\"* Then follow the relevant workflow below.\n\n---\n\n## Step 1 — Choose your index configuration\n\nPick the index type before writing any code. Getting this wrong means\nre-creating the index later.\n\n**Serverless (recommended for most cases)**\n```python\nfrom pinecone import Pinecone, ServerlessSpec\n\npc = Pinecone(api_key=\"PINECONE_API_KEY\")\n\nif \"my-index\" not in pc.list_indexes().names():\n    pc.create_index(\n        name=\"my-index\",\n        dimension=1536,        # must match your embedding model exactly\n        metric=\"cosine\",\n        spec=ServerlessSpec(cloud=\"aws\", region=\"us-east-1\")\n    )\nindex = pc.Index(\"my-index\")\n```\n\n**Pod-based (for consistent high-throughput production)**\n```python\nfrom pinecone import PodSpec\n\npc.create_index(\n    name=\"my-index-prod\",\n    dimension=1536,\n    metric=\"cosine\",\n    spec=PodSpec(environment=\"us-east1-gcp\", pod_type=\"p1.x1\")\n)\n```\n\n**Dimension quick reference — match this exactly to your embedding model:**\n| Model | Dimension |\n|---|---|\n| `text-embedding-3-small` | 1536 |\n| `text-embedding-3-large` | 3072 |\n| `voyage-3` \u002F `voyage-multimodal-3` | 1024 |\n| `BAAI\u002Fbge-large-en-v1.5` | 1024 |\n| `intfloat\u002Fmultilingual-e5-large` (Arabic, Malay, Chinese) | 1024 |\n\n> **Checkpoint**: Index exists, dimension matches embedding model, `index.describe_index_stats()` returns without error.\n\n---\n\n## Step 2 — Embed and upsert documents\n\nAlways batch upserts — never upsert one vector at a time.\n\n```python\nfrom openai import OpenAI\n\nclient = OpenAI()\n\ndef embed(texts: list[str]) -> list[list[float]]:\n    res = client.embeddings.create(model=\"text-embedding-3-small\", input=texts)\n    return [r.embedding for r in res.data]\n\ndef upsert_docs(index, docs: list[dict], namespace: str = \"default\"):\n    \"\"\"docs = [{\"id\": \"...\", \"text\": \"...\", \"metadata\": {...}}]\"\"\"\n    BATCH = 100\n    for i in range(0, len(docs), BATCH):\n        batch = docs[i:i + BATCH]\n        vecs = [\n            {\n                \"id\": d[\"id\"],\n                \"values\": emb,\n                \"metadata\": {**d.get(\"metadata\", {}), \"text\": d[\"text\"]}\n            }\n            for d, emb in zip(batch, embed([d[\"text\"] for d in batch]))\n        ]\n        index.upsert(vectors=vecs, namespace=namespace)\n```\n\n**Always store the original text in metadata** — this avoids a second lookup\nat retrieval time.\n\n> **Checkpoint**: `index.describe_index_stats()` shows vector count > 0 in the\n> target namespace.\n\n---\n\n## Step 3 — Choose retrieval strategy\n\n### Dense (semantic) search — use for most cases\n```python\ndef search(index, query: str, top_k: int = 5, namespace: str = \"default\",\n           filter: dict = None) -> list[dict]:\n    [q_emb] = embed([query])\n    results = index.query(\n        vector=q_emb, top_k=top_k, namespace=namespace,\n        include_metadata=True, filter=filter\n    )\n    return [{\"text\": m.metadata[\"text\"], \"score\": m.score, \"id\": m.id}\n            for m in results.matches]\n```\n\n### Hybrid search (semantic + BM25 keyword) — use when corpus has exact terminology\nUse hybrid when the domain has precise terms that semantic search misses:\nlegal citations, medical codes, product SKUs, API method names.\n\n```python\nfrom pinecone_text.sparse import BM25Encoder\n\nbm25 = BM25Encoder().default()\nbm25.fit([d[\"text\"] for d in docs])  # fit once on your corpus\n\ndef hybrid_search(index, query: str, top_k: int = 5, alpha: float = 0.7):\n    \"\"\"alpha=1.0 is pure dense; alpha=0.0 is pure sparse.\"\"\"\n    dense = [v * alpha for v in embed([query])[0]]\n    sparse_raw = bm25.encode_queries(query)\n    sparse = {\n        \"indices\": sparse_raw[\"indices\"],\n        \"values\": [v * (1 - alpha) for v in sparse_raw[\"values\"]]\n    }\n    return index.query(vector=dense, sparse_vector=sparse,\n                       top_k=top_k, include_metadata=True).matches\n```\n\n### Metadata filtering — use to scope results before semantic ranking\n```python\n# Exact match\nresults = index.query(vector=emb, filter={\"source\": {\"$eq\": \"confluence\"}})\n\n# Combined filter\nresults = index.query(vector=emb, filter={\n    \"$and\": [\n        {\"category\": {\"$eq\": \"engineering\"}},\n        {\"language\": {\"$in\": [\"en\", \"ar\"]}}\n    ]\n})\n```\n\n> **Checkpoint**: A test query returns relevant results with scores > 0.7 for\n> clearly matching content.\n\n---\n\n## Step 4A — Full RAG pipeline (document use case)\n\n```python\ndef rag_answer(index, question: str, namespace: str = \"default\",\n               model: str = \"gpt-4o-mini\") -> str:\n    hits = search(index, question, top_k=5, namespace=namespace)\n    context = \"\\n\\n\".join(h[\"text\"] for h in hits)\n\n    return client.chat.completions.create(\n        model=model,\n        messages=[\n            {\n                \"role\": \"system\",\n                \"content\": (\n                    \"Answer using only the provided context. \"\n                    \"If the answer isn't in the context, say so.\\n\\n\"\n                    f\"Context:\\n{context}\"\n                )\n            },\n            {\"role\": \"user\", \"content\": question}\n        ]\n    ).choices[0].message.content\n```\n\n---\n\n## Step 4B — Agent memory (memory use case)\n\nUse namespaces to isolate each agent's or user's memories completely.\nNamespace per agent prevents memory bleed across users or sessions.\n\n```python\nimport time, hashlib\n\ndef remember(index, agent_id: str, content: str,\n             memory_type: str = \"fact\"):\n    \"\"\"Store a memory for an agent.\"\"\"\n    mem_id = hashlib.md5(\n        f\"{agent_id}{content}{time.time()}\".encode()\n    ).hexdigest()\n    [emb] = embed([content])\n    index.upsert(\n        vectors=[{\n            \"id\": mem_id,\n            \"values\": emb,\n            \"metadata\": {\n                \"text\": content,\n                \"type\": memory_type,\n                \"timestamp\": time.time(),\n                \"agent_id\": agent_id\n            }\n        }],\n        namespace=f\"agent_{agent_id}\"\n    )\n\ndef recall(index, agent_id: str, query: str,\n           top_k: int = 5) -> list[str]:\n    \"\"\"Recall relevant memories for an agent.\"\"\"\n    return [h[\"text\"] for h in\n            search(index, query, top_k=top_k,\n                   namespace=f\"agent_{agent_id}\")]\n\ndef forget(index, agent_id: str):\n    \"\"\"Wipe all memories for an agent (e.g., on user request).\"\"\"\n    index.delete(delete_all=True, namespace=f\"agent_{agent_id}\")\n```\n\n---\n\n## Step 5 — Wire it together and test end to end\n\nRun a quick smoke test before integrating into the larger system:\n\n```python\n# Smoke test\nupsert_docs(index, [\n    {\"id\": \"t1\", \"text\": \"Pinecone is a vector database for semantic search.\"},\n    {\"id\": \"t2\", \"text\": \"RAG combines retrieval with language model generation.\"},\n])\n\nhits = search(index, \"What is Pinecone?\")\nassert hits[0][\"score\"] > 0.7, f\"Expected high similarity, got {hits[0]['score']}\"\nprint(\"Smoke test passed:\", hits[0][\"text\"])\n```\n\n> **Checkpoint**: Smoke test passes. End-to-end: index → upsert → query →\n> LLM response works without errors.\n\n---\n\n## Common pitfalls — fix these before they become bugs\n\n- **Dimension mismatch**: always verify `len(embed([\"test\"])[0])` matches\n  the index dimension before your first upsert.\n- **Missing text in metadata**: if you don't store `\"text\"` in metadata,\n  you'll need a second lookup to get the actual content at query time.\n- **Single-vector upserts in a loop**: always batch in chunks of 100.\n- **No namespace strategy**: decide upfront — one namespace per user\u002Fagent\n  prevents cross-tenant data leaks that are hard to fix later.\n- **Fitting BM25 on a small corpus**: BM25 needs a representative corpus to\n  build good term frequencies. Fit on at least a few hundred documents.\n\n## When NOT to use this skill\n\nUse a different approach when:\n- The dataset fits in memory and latency doesn't matter → use FAISS or Chroma\n- You're already on PostgreSQL and want to avoid a new service → use pgvector\n- You need sub-5ms p99 latency with no external API calls → local vector store\n- The user explicitly wants a different vector DB (Weaviate, Qdrant, etc.)",{"name":5,"license":15,"description":60,"compatibility":61},"Build production RAG pipelines and persistent agent memory using Pinecone as the vector database backend. ALWAYS USE THIS SKILL when the user mentions Pinecone, wants to index documents for semantic search, build a retrieval-augmented generation system, store agent memory across sessions, implement hybrid search, or connect an LLM to a searchable knowledge base — even if they don't say \"Pinecone\" explicitly. Also use when the user asks about vector databases for RAG, namespace isolation for multi-tenant agents, embedding pipelines, or scaling a knowledge base beyond what local storage can handle. DO NOT use for local-only vector stores (Chroma, FAISS, pgvector) or pure keyword search with no semantic component.\n","pinecone>=6.0.0, Python 3.10+","main","5269f7ea2d6c3a9bfda5ed878b6726cf509121fd","skills\u002Fpinecone-rag",5340,"https:\u002F\u002Fgithub.com\u002Fgithub\u002Fawesome-copilot","https:\u002F\u002Fawesome-copilot.github.com\u002F","https:\u002F\u002Fgithub.com\u002Fgithub","https:\u002F\u002Fgithub.com\u002Fabout","How people build software.","2026-11-11 13:41:32",{"source":8,"skillName":5,"commands":73,"prompt":83,"zipAvailable":84},[74,77,80],{"tool":75,"command":76},"npx","npx skills add github\u002Fawesome-copilot --skill pinecone-rag",{"tool":78,"command":79},"bunx","bunx skills add github\u002Fawesome-copilot --skill pinecone-rag",{"tool":81,"command":82},"pnpm","pnpm dlx skills add github\u002Fawesome-copilot --skill pinecone-rag","安装技能，技能地址：https:\u002F\u002Fgithub.com\u002Fgithub\u002Fawesome-copilot\u002Fblob\u002Fmain\u002Fskills\u002Fpinecone-rag\u002FSKILL.md",true,[86,92,97],{"path":87,"resourceType":88,"resourceTypeDesc":89,"sizeBytes":90,"sourceHtmlUrl":91},".gitignore","OTHER","其他",41,"https:\u002F\u002Fgithub.com\u002Fshengdabai\u002FTony-Claude-Code-Skills\u002Fblob\u002Fmain\u002Fskills\u002Fhumanizer-zh\u002F.gitignore",{"path":93,"resourceType":93,"resourceTypeDesc":94,"sizeBytes":95,"sourceHtmlUrl":96},"LICENSE","许可证",1063,"https:\u002F\u002Fgithub.com\u002Fshengdabai\u002FTony-Claude-Code-Skills\u002Fblob\u002Fmain\u002Fskills\u002Fhumanizer-zh\u002FLICENSE",{"path":98,"resourceType":99,"resourceTypeDesc":100,"sizeBytes":101,"sourceHtmlUrl":102},"README.md","REFERENCE","参考文档",7799,"https:\u002F\u002Fgithub.com\u002Fshengdabai\u002FTony-Claude-Code-Skills\u002Fblob\u002Fmain\u002Fskills\u002Fhumanizer-zh\u002FREADME.md",[104,135,163,192,223,239],{"skillNo":105,"name":106,"nameZh":6,"description":107,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":108,"licenseFileUrl":6,"sourceHtmlUrl":109,"categories":110,"keywords":115,"publishedAt":55},"01M70KXZNX9ES38N2A8M58ZZMZ","supabase-rls-policies","Write, review, and test Supabase Row Level Security (RLS) policies so that one user cannot read or change another user's rows. Use when creating or changing tables, policies, views, or security definer functions in a Supabase project, when reviewing a migration that touches RLS, or when asked to prove user isolation with pgTAP tests that sign in as two users.","MIT","https:\u002F\u002Fgithub.com\u002Fgithub\u002Fawesome-copilot\u002Fblob\u002Fmain\u002Fskills\u002Fsupabase-rls-policies\u002FSKILL.md",[111,114],{"categoryNo":112,"categoryName":113,"taxonomyType":21,"taxonomyTypeDesc":22},"01JBSEEDCAT000000000000008","测试与质量",{"categoryNo":24,"categoryName":25,"taxonomyType":21,"taxonomyTypeDesc":22},[116,117,118,119,120,31,121,122,123,124,125,126,127,128,129,130,131,132,133,134],"supabase","rls","supabase rls","supabase rls policies","project","review","change","read","write","security","isolation","tables","as","tests","prove","use","users","functions","with",{"skillNo":136,"name":137,"nameZh":6,"description":138,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":108,"licenseFileUrl":6,"sourceHtmlUrl":139,"categories":140,"keywords":142,"publishedAt":55},"01M70KWA5YWXDV6V8DZ19E3T2Y","skill-distribution-audit","Audit a skill collection before release or investigate installer omissions, archived\u002Ftemplate leakage, duplicate identities, and manifest drift. Reconcile the intended public source set, distribution manifest, and actual discovery output for a specified consumer version. Includes a deterministic set-comparison helper.","https:\u002F\u002Fgithub.com\u002Fgithub\u002Fawesome-copilot\u002Fblob\u002Fmain\u002Fskills\u002Fskill-distribution-audit\u002FSKILL.md",[141],{"categoryNo":24,"categoryName":25,"taxonomyType":21,"taxonomyTypeDesc":22},[143,144,145,146,147,148,149,150,151,152,153,154,155,156,157,158,159,160,161,162],"distribution","skill distribution audit","audit","SET","deterministic","installer","for","discovery","comparison","helper","manifest","version","collection","drift","output","investigate","duplicate","template","release","reconcile",{"skillNo":164,"name":165,"nameZh":6,"description":166,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":108,"licenseFileUrl":167,"sourceHtmlUrl":168,"categories":169,"keywords":175,"publishedAt":55},"01M70KCJPFVW7PPQ49CC5R8JGK","logicapps-custom-code-net10-migration","Migrate Azure Logic Apps Standard local-function custom-code projects from .NET 8 to .NET 10. Use for legacy Logic Apps custom-code workspace upgrades, project\u002Fpackage configuration, custom-worker settings, publication, and integrated VS Code debugging. Not for ordinary .NET applications, standalone Azure Functions, Consumption workflows, or cloud deployment.","https:\u002F\u002Fgithub.com\u002Fgithub\u002Fawesome-copilot\u002Fblob\u002Fmain\u002Fskills\u002Flogicapps-custom-code-net10-migration\u002FLICENSE","https:\u002F\u002Fgithub.com\u002Fgithub\u002Fawesome-copilot\u002Fblob\u002Fmain\u002Fskills\u002Flogicapps-custom-code-net10-migration\u002FSKILL.md",[170],{"categoryNo":171,"categoryName":172,"taxonomyType":173,"taxonomyTypeDesc":174},"01JBSEEDCAT000000000000030","开发工程师","PROFESSION","适用职业",[176,177,178,179,180,181,149,120,182,183,36,184,185,186,187,131,188,189,190,191],"custom","code","logicapps custom code net10 migration","migration","projects","standalone","deployment","applications","azure","worker","upgrades","package","apps","Legacy","debugging","workflows",{"skillNo":193,"name":194,"nameZh":6,"description":195,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":108,"licenseFileUrl":196,"sourceHtmlUrl":197,"categories":198,"keywords":201,"publishedAt":55},"01M6AEBN37RQMXJ3298PEY7PG1","winmd-api-search","Find and explore Windows desktop APIs. Use when building features that need platform capabilities — camera, file access, notifications, UI controls, AI\u002FML, sensors, networking, etc. Discovers the right API for a task and retrieves full type details (methods, properties, events, enumeration values).","https:\u002F\u002Fgithub.com\u002Fgithub\u002Fawesome-copilot\u002Fblob\u002Fmain\u002Fskills\u002Fwinmd-api-search\u002FLICENSE.txt","https:\u002F\u002Fgithub.com\u002Fgithub\u002Fawesome-copilot\u002Fblob\u002Fmain\u002Fskills\u002Fwinmd-api-search\u002FSKILL.md",[199,200],{"categoryNo":19,"categoryName":20,"taxonomyType":21,"taxonomyTypeDesc":22},{"categoryNo":171,"categoryName":172,"taxonomyType":173,"taxonomyTypeDesc":174},[202,43,203,204,205,206,207,208,209,210,211,212,213,214,215,31,216,217,218,219,220,149,221,222],"winmd api search","api","windows","desktop","file access","enumeration","ui","ml","file","networking","capabilities","notifications","explore","access","features","full","camera","events","platform","methods","controls",{"skillNo":224,"name":225,"nameZh":6,"description":226,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":227,"sourceHtmlUrl":228,"categories":229,"keywords":235,"publishedAt":55},"01M6AE571R3SC1ZGE202B3NB5S","resemble-detect","Deepfake detection and media safety — detect AI-generated audio, images, video, and text, trace synthesis sources, apply watermarks, verify speaker identity, and analyze media intelligence using Resemble AI","https:\u002F\u002Fgithub.com\u002Fgithub\u002Fawesome-copilot\u002Fblob\u002Fmain\u002Fskills\u002Fresemble-detect\u002FLICENSE","https:\u002F\u002Fgithub.com\u002Fgithub\u002Fawesome-copilot\u002Fblob\u002Fmain\u002Fskills\u002Fresemble-detect\u002FSKILL.md",[230,231,234],{"categoryNo":19,"categoryName":20,"taxonomyType":21,"taxonomyTypeDesc":22},{"categoryNo":232,"categoryName":233,"taxonomyType":21,"taxonomyTypeDesc":22},"01JBSEEDCAT000000000000016","图像与多媒体",{"categoryNo":24,"categoryName":25,"taxonomyType":21,"taxonomyTypeDesc":22},[236,146,210,237,238],"resemble detect","https","identity",{"skillNo":240,"name":241,"nameZh":6,"description":242,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":243,"sourceHtmlUrl":244,"categories":245,"keywords":250,"publishedAt":55},"01M6AE30BSK2HCY4TQQ22Z8JRD","quality-playbook","Run a complete quality engineering audit on any codebase. Derives behavioral requirements from the code, generates spec-traced functional tests, runs a three-pass code review with regression tests, executes a multi-model spec audit (Council of Three), and produces a consolidated bug report with TDD-verified patches. Finds the 35% of real defects that structural code review alone cannot catch. Works with any language. Trigger on 'quality playbook', 'spec audit', 'Council of Three', 'fitness-to-purpose', or 'coverage theater'.","https:\u002F\u002Fgithub.com\u002Fgithub\u002Fawesome-copilot\u002Fblob\u002Fmain\u002Fskills\u002Fquality-playbook\u002FLICENSE.txt","https:\u002F\u002Fgithub.com\u002Fgithub\u002Fawesome-copilot\u002Fblob\u002Fmain\u002Fskills\u002Fquality-playbook\u002FSKILL.md",[246,249],{"categoryNo":247,"categoryName":248,"taxonomyType":21,"taxonomyTypeDesc":22},"01JBSEEDCAT000000000000007","代码审查",{"categoryNo":24,"categoryName":25,"taxonomyType":21,"taxonomyTypeDesc":22},[251,252,253,254,177,121,9,145,255,256,237,257,258,259,260,261,262,263,264,265,266,267,129,268,269,270,271],"playbook","quality playbook","quality","code review","tdd","run","report","spec","bug report","council","engineering","functional","spec audit","bug","verified","multi-model","three","coverage","model","regression","structural"]