# Send Langsearch to your agent Hand the extracted package to your coding agent with a concrete install brief instead of figuring it out manually. ## Fast path - 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. ## Suggested prompts ### New install ```text 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. ``` ### Upgrade existing ```text 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. ``` ## Machine-readable fields ```json { "schemaVersion": "1.0", "item": { "slug": "langsearch", "name": "Langsearch", "source": "tencent", "type": "skill", "category": "开发工具", "sourceUrl": "https://clawhub.ai/vaibhav1805/langsearch", "canonicalUrl": "https://clawhub.ai/vaibhav1805/langsearch", "targetPlatform": "OpenClaw" }, "install": { "downloadUrl": "/downloads/langsearch", "sourceDownloadUrl": "https://wry-manatee-359.convex.site/api/v1/download?slug=langsearch", "sourcePlatform": "tencent", "targetPlatform": "OpenClaw", "packageFormat": "ZIP package", "primaryDoc": "SKILL.md", "includedAssets": [ "scripts/web_search_example.py", "SKILL.md" ], "downloadMode": "redirect", "sourceHealth": { "source": "tencent", "slug": "langsearch", "status": "healthy", "reason": "direct_download_ok", "recommendedAction": "download", "checkedAt": "2026-05-04T20:43:57.058Z", "expiresAt": "2026-05-11T20:43:57.058Z", "httpStatus": 200, "finalUrl": "https://wry-manatee-359.convex.site/api/v1/download?slug=langsearch", "contentType": "application/zip", "probeMethod": "head", "details": { "probeUrl": "https://wry-manatee-359.convex.site/api/v1/download?slug=langsearch", "contentDisposition": "attachment; filename=\"langsearch-1.0.4.zip\"", "redirectLocation": null, "bodySnippet": null, "slug": "langsearch" }, "scope": "item", "summary": "Item download looks usable.", "detail": "Yavira can redirect you to the upstream package for this item.", "primaryActionLabel": "Download for OpenClaw", "primaryActionHref": "/downloads/langsearch" }, "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." ] } }, "links": { "detailUrl": "https://openagent3.xyz/skills/langsearch", "downloadUrl": "https://openagent3.xyz/downloads/langsearch", "agentUrl": "https://openagent3.xyz/skills/langsearch/agent", "manifestUrl": "https://openagent3.xyz/skills/langsearch/agent.json", "briefUrl": "https://openagent3.xyz/skills/langsearch/agent.md" } } ``` ## Documentation ### ⚠️ Security & Credentials Required Credentials: LANGSEARCH_API_KEY - Your free LangSearch API key (required) Security Best Practices: Get a free API key - Sign up at langsearch.com/api-keys Protect your API key - Never commit .env files containing LANGSEARCH_API_KEY to version control Use environment variables - Store the key in .env or set via export LANGSEARCH_API_KEY="..." Monitor usage - Check your API usage on the LangSearch dashboard Code inspection - This tool only uses the official LangSearch API. All communication is via HTTPS to api.langsearch.com Network Access: Only connects to: https://api.langsearch.com (official LangSearch API) No external data collection or telemetry No tracking or logging sent elsewhere ### Overview LangSearch provides free APIs for web search and semantic reranking. It combines keyword search precision with vector-based semantic matching, making it ideal for integrating current web information into LLM applications and building RAG systems. ### Prerequisites Get a free API key at https://langsearch.com/api-keys Set your API key as an environment variable: export LANGSEARCH_API_KEY="your-api-key" ### Basic Web Search The simplest way to search the web using LangSearch: import requests import json import os api_key = os.getenv("LANGSEARCH_API_KEY") headers = { "Authorization": f"Bearer {api_key}", "Content-Type": "application/json" } query = "latest AI developments 2025" payload = { "query": query, "count": 5, "summary": True } response = requests.post( "https://api.langsearch.com/v1/web-search", headers=headers, json=payload ) results = response.json() print(json.dumps(results, indent=2)) Or using cURL: curl -X POST https://api.langsearch.com/v1/web-search \\ -H "Authorization: Bearer $LANGSEARCH_API_KEY" \\ -H "Content-Type: application/json" \\ -d '{ "query": "latest AI developments 2025", "count": 5, "summary": true }' ### Web Search API The web search endpoint retrieves information from billions of web documents using hybrid search (keyword + vector matching) with optional summaries. ### Endpoint POST https://api.langsearch.com/v1/web-search ### Request Parameters ParameterTypeRequiredDescriptionquerystringYesThe search querycountintegerNoNumber of results to return (default: 10, max: 100)summarybooleanNoInclude markdown summaries in results (default: false)freshnessstringNoFilter results by freshness (day, week, month, year) ### Response Structure The API returns an array of search results with: title - Result title url - Result URL snippet - Text excerpt from the page summary - Markdown formatted summary (if summary: true in request) score - Relevance score ### Example Use Cases Current Information Retrieval When your LLM needs up-to-date information: # Search for recent developments response = requests.post( "https://api.langsearch.com/v1/web-search", headers=headers, json={ "query": "Python 3.14 release notes", "count": 3, "summary": True, "freshness": "week" } ) Multi-Query Research Build context from multiple searches: queries = [ "climate change mitigation strategies", "renewable energy trends 2025", "carbon capture technology" ] all_results = [] for query in queries: response = requests.post( "https://api.langsearch.com/v1/web-search", headers=headers, json={"query": query, "count": 5, "summary": True} ) all_results.extend(response.json()) ### Semantic Reranking API Improve search accuracy by reranking results based on semantic relevance to your query. ### Endpoint POST https://api.langsearch.com/v1/rerank ### Request Parameters ParameterTypeRequiredDescriptionquerystringYesThe search query for contextdocumentsarrayYesArray of documents to rerank (each with title, text, etc.)modelstringNoReranking model (default: langsearch-rerank-1)top_nintegerNoNumber of top results to return (default: 10)return_documentsbooleanNoInclude full documents in response (default: false) ### Example: Reranking Web Search Results # Get initial search results search_response = requests.post( "https://api.langsearch.com/v1/web-search", headers=headers, json={"query": "machine learning deployment best practices", "count": 10} ) search_results = search_response.json() # Prepare documents for reranking documents = [ {"title": r.get("title", ""), "text": r.get("snippet", "")} for r in search_results ] # Rerank for better relevance rerank_response = requests.post( "https://api.langsearch.com/v1/rerank", headers=headers, json={ "query": "best practices for deploying ML models in production", "documents": documents, "top_n": 5 } ) reranked = rerank_response.json() ### Building RAG Applications Combine web search with LLM context for better information retrieval and generation: def rag_query(user_question): # Step 1: Search the web for relevant information search_response = requests.post( "https://api.langsearch.com/v1/web-search", headers=headers, json={ "query": user_question, "count": 5, "summary": True } ) search_results = search_response.json() # Step 2: Extract summaries and URLs for context context = "\\n".join([ f"- {r['title']}: {r.get('summary', r.get('snippet', ''))}" for r in search_results ]) # Step 3: Use with your LLM # This is where you'd call your LLM with the context rag_context = f""" Based on recent web search results: {context} Answer the user's question: {user_question} """ return rag_context, search_results ### Error Handling Common HTTP status codes: StatusMeaningAction200SuccessResults returned normally401UnauthorizedCheck API key is valid and set correctly429Rate limitedRetry with exponential backoff500Server errorRetry the request later See scripts/web_search_example.py for a complete example with error handling. ### Resources This skill includes example resource directories that demonstrate how to organize different types of bundled resources: ### scripts/ Executable code (Python/Bash/etc.) that can be run directly to perform specific operations. Examples from other skills: PDF skill: fill_fillable_fields.py, extract_form_field_info.py - utilities for PDF manipulation DOCX skill: document.py, utilities.py - Python modules for document processing Appropriate for: Python scripts, shell scripts, or any executable code that performs automation, data processing, or specific operations. Note: Scripts may be executed without loading into context, but can still be read by Claude for patching or environment adjustments. ### references/ Documentation and reference material intended to be loaded into context to inform Claude's process and thinking. Examples from other skills: Product management: communication.md, context_building.md - detailed workflow guides BigQuery: API reference documentation and query examples Finance: Schema documentation, company policies Appropriate for: In-depth documentation, API references, database schemas, comprehensive guides, or any detailed information that Claude should reference while working. ### assets/ Files not intended to be loaded into context, but rather used within the output Claude produces. Examples from other skills: Brand styling: PowerPoint template files (.pptx), logo files Frontend builder: HTML/React boilerplate project directories Typography: Font files (.ttf, .woff2) Appropriate for: Templates, boilerplate code, document templates, images, icons, fonts, or any files meant to be copied or used in the final output. Any unneeded directories can be deleted. Not every skill requires all three types of resources. ## Trust - Source: tencent - Verification: Indexed source record - Publisher: vaibhav1805 - Version: 1.0.4 ## Source health - Status: healthy - Item download looks usable. - Yavira can redirect you to the upstream package for this item. - Health scope: item - Reason: direct_download_ok - Checked at: 2026-05-04T20:43:57.058Z - Expires at: 2026-05-11T20:43:57.058Z - Recommended action: Download for OpenClaw ## Links - [Detail page](https://openagent3.xyz/skills/langsearch) - [Send to Agent page](https://openagent3.xyz/skills/langsearch/agent) - [JSON manifest](https://openagent3.xyz/skills/langsearch/agent.json) - [Markdown brief](https://openagent3.xyz/skills/langsearch/agent.md) - [Download page](https://openagent3.xyz/downloads/langsearch)