{ "schemaVersion": "1.0", "item": { "slug": "embeddings", "name": "Embeddings", "source": "tencent", "type": "skill", "category": "开发工具", "sourceUrl": "https://clawhub.ai/ivangdavila/embeddings", "canonicalUrl": "https://clawhub.ai/ivangdavila/embeddings", "targetPlatform": "OpenClaw" }, "install": { "downloadMode": "redirect", "downloadUrl": "/downloads/embeddings", "sourceDownloadUrl": "https://wry-manatee-359.convex.site/api/v1/download?slug=embeddings", "sourcePlatform": "tencent", "targetPlatform": "OpenClaw", "installMethod": "Manual import", "extraction": "Extract archive", "prerequisites": [ "OpenClaw" ], "packageFormat": "ZIP package", "includedAssets": [ "SKILL.md", "chunking.md", "providers.md", "search.md", "storage.md" ], "primaryDoc": "SKILL.md", "quickSetup": [ "Download the package from Yavira.", "Extract the archive and review SKILL.md first.", "Import or place the package into your OpenClaw setup." ], "agentAssist": { "summary": "Hand the extracted package to your coding agent with a concrete install brief instead of figuring it out manually.", "steps": [ "Download the package from Yavira.", "Extract it into a folder your agent can access.", "Paste one of the prompts below and point your agent at the extracted folder." ], "prompts": [ { "label": "New install", "body": "I downloaded a skill package from Yavira. Read SKILL.md from the extracted folder and install it by following the included instructions. Tell me what you changed and call out any manual steps you could not complete." }, { "label": "Upgrade existing", "body": "I downloaded an updated skill package from Yavira. Read SKILL.md from the extracted folder, compare it with my current installation, and upgrade it while preserving any custom configuration unless the package docs explicitly say otherwise. Summarize what changed and any follow-up checks I should run." } ] }, "sourceHealth": { "source": "tencent", "status": "healthy", "reason": "direct_download_ok", "recommendedAction": "download", "checkedAt": "2026-04-30T16:55:25.780Z", "expiresAt": "2026-05-07T16:55:25.780Z", "httpStatus": 200, "finalUrl": "https://wry-manatee-359.convex.site/api/v1/download?slug=network", "contentType": "application/zip", "probeMethod": "head", "details": { "probeUrl": "https://wry-manatee-359.convex.site/api/v1/download?slug=network", "contentDisposition": "attachment; filename=\"network-1.0.0.zip\"", "redirectLocation": null, "bodySnippet": null }, "scope": "source", "summary": "Source download looks usable.", "detail": "Yavira can redirect you to the upstream package for this source.", "primaryActionLabel": "Download for OpenClaw", "primaryActionHref": "/downloads/embeddings" }, "validation": { "installChecklist": [ "Use the Yavira download entry.", "Review SKILL.md after the package is downloaded.", "Confirm the extracted package contains the expected setup assets." ], "postInstallChecks": [ "Confirm the extracted package includes the expected docs or setup files.", "Validate the skill or prompts are available in your target agent workspace.", "Capture any manual follow-up steps the agent could not complete." ] }, "downloadPageUrl": "https://openagent3.xyz/downloads/embeddings", "agentPageUrl": "https://openagent3.xyz/skills/embeddings/agent", "manifestUrl": "https://openagent3.xyz/skills/embeddings/agent.json", "briefUrl": "https://openagent3.xyz/skills/embeddings/agent.md" }, "agentAssist": { "summary": "Hand the extracted package to your coding agent with a concrete install brief instead of figuring it out manually.", "steps": [ "Download the package from Yavira.", "Extract it into a folder your agent can access.", "Paste one of the prompts below and point your agent at the extracted folder." ], "prompts": [ { "label": "New install", "body": "I downloaded a skill package from Yavira. Read SKILL.md from the extracted folder and install it by following the included instructions. Tell me what you changed and call out any manual steps you could not complete." }, { "label": "Upgrade existing", "body": "I downloaded an updated skill package from Yavira. Read SKILL.md from the extracted folder, compare it with my current installation, and upgrade it while preserving any custom configuration unless the package docs explicitly say otherwise. Summarize what changed and any follow-up checks I should run." } ] }, "documentation": { "source": "clawhub", "primaryDoc": "SKILL.md", "sections": [ { "title": "When to Use", "body": "User wants to convert text/images to vectors, build semantic search, or integrate embeddings into applications." }, { "title": "Quick Reference", "body": "TopicFileProvider comparison & selectionproviders.mdChunking strategies & codechunking.mdVector database patternsstorage.mdSearch & retrieval tuningsearch.md" }, { "title": "Core Capabilities", "body": "Generate embeddings — Call provider APIs (OpenAI, Cohere, Voyage, local models)\nChunk content — Split documents with overlap, semantic boundaries, token limits\nStore vectors — Insert into Pinecone, Weaviate, Qdrant, pgvector, Chroma\nSimilarity search — Query with top-k, filters, hybrid search\nBatch processing — Handle large datasets with rate limiting and retries\nModel comparison — Evaluate embedding quality for specific use cases" }, { "title": "Decision Checklist", "body": "Before recommending approach, ask:\n\nWhat content type? (text, code, images, multimodal)\n Volume and update frequency?\n Latency requirements? (real-time vs batch)\n Budget constraints? (API costs vs self-hosted)\n Existing infrastructure? (cloud provider, database)" }, { "title": "Critical Rules", "body": "Same model everywhere — Query embeddings MUST use identical model as document embeddings\nNormalize before storage — Most similarity metrics assume unit vectors\nChunk with overlap — 10-20% overlap prevents context loss at boundaries\nBatch API calls — Never embed one item at a time in production\nCache embeddings — Regenerating is expensive; store with source hash\nMonitor dimensions — Higher isn't always better; 768-1536 is usually optimal" }, { "title": "Provider Quick Selection", "body": "NeedProviderWhyBest quality, any costOpenAI text-embedding-3-largeTop benchmarksCost-sensitiveOpenAI text-embedding-3-small5x cheaper, 80% qualityMultilingualCohere embed-multilingual-v3100+ languagesCode/technicalVoyage voyage-code-2Optimized for codePrivacy/offlineLocal (e5, bge, nomic)No data leaves machineImagesOpenAI CLIP, Cohere multimodalCross-modal search" }, { "title": "Common Patterns", "body": "# Batch embedding with retry\ndef embed_batch(texts, model=\"text-embedding-3-small\"):\n results = []\n for chunk in batched(texts, 100): # API limit\n response = client.embeddings.create(input=chunk, model=model)\n results.extend([e.embedding for e in response.data])\n return results\n\n# Similarity search with filter\nresults = index.query(\n vector=query_embedding,\n top_k=10,\n filter={\"category\": \"technical\"},\n include_metadata=True\n)" } ], "body": "When to Use\n\nUser wants to convert text/images to vectors, build semantic search, or integrate embeddings into applications.\n\nQuick Reference\nTopic\tFile\nProvider comparison & selection\tproviders.md\nChunking strategies & code\tchunking.md\nVector database patterns\tstorage.md\nSearch & retrieval tuning\tsearch.md\nCore Capabilities\nGenerate embeddings — Call provider APIs (OpenAI, Cohere, Voyage, local models)\nChunk content — Split documents with overlap, semantic boundaries, token limits\nStore vectors — Insert into Pinecone, Weaviate, Qdrant, pgvector, Chroma\nSimilarity search — Query with top-k, filters, hybrid search\nBatch processing — Handle large datasets with rate limiting and retries\nModel comparison — Evaluate embedding quality for specific use cases\nDecision Checklist\n\nBefore recommending approach, ask:\n\n What content type? (text, code, images, multimodal)\n Volume and update frequency?\n Latency requirements? (real-time vs batch)\n Budget constraints? (API costs vs self-hosted)\n Existing infrastructure? (cloud provider, database)\nCritical Rules\nSame model everywhere — Query embeddings MUST use identical model as document embeddings\nNormalize before storage — Most similarity metrics assume unit vectors\nChunk with overlap — 10-20% overlap prevents context loss at boundaries\nBatch API calls — Never embed one item at a time in production\nCache embeddings — Regenerating is expensive; store with source hash\nMonitor dimensions — Higher isn't always better; 768-1536 is usually optimal\nProvider Quick Selection\nNeed\tProvider\tWhy\nBest quality, any cost\tOpenAI text-embedding-3-large\tTop benchmarks\nCost-sensitive\tOpenAI text-embedding-3-small\t5x cheaper, 80% quality\nMultilingual\tCohere embed-multilingual-v3\t100+ languages\nCode/technical\tVoyage voyage-code-2\tOptimized for code\nPrivacy/offline\tLocal (e5, bge, nomic)\tNo data leaves machine\nImages\tOpenAI CLIP, Cohere multimodal\tCross-modal search\nCommon Patterns\n# Batch embedding with retry\ndef embed_batch(texts, model=\"text-embedding-3-small\"):\n results = []\n for chunk in batched(texts, 100): # API limit\n response = client.embeddings.create(input=chunk, model=model)\n results.extend([e.embedding for e in response.data])\n return results\n\n# Similarity search with filter\nresults = index.query(\n vector=query_embedding,\n top_k=10,\n filter={\"category\": \"technical\"},\n include_metadata=True\n)" }, "trust": { "sourceLabel": "tencent", "provenanceUrl": "https://clawhub.ai/ivangdavila/embeddings", "publisherUrl": "https://clawhub.ai/ivangdavila/embeddings", "owner": "ivangdavila", "version": "1.0.0", "license": null, "verificationStatus": "Indexed source record" }, "links": { "detailUrl": "https://openagent3.xyz/skills/embeddings", "downloadUrl": "https://openagent3.xyz/downloads/embeddings", "agentUrl": "https://openagent3.xyz/skills/embeddings/agent", "manifestUrl": "https://openagent3.xyz/skills/embeddings/agent.json", "briefUrl": "https://openagent3.xyz/skills/embeddings/agent.md" } }