LangSearch for Openclaw

A comprehensive API for integrating free web search and semantic reranking into AGI applications and RAG pipelines.

vaibhav1805
v1.0.4
Feb 25, 2026
2
1.5k
0

Install & Download

1. ClawHub CLI

The fastest way to install a skill directly from the registry.

npx clawhub@latest install langsearch

2. Manual Installation

Copy the skill folder to one of these locations

Global
~/.openclaw/skills/
Workspace
<project>/skills/

Priority: Workspace > Local > Bundled

3. Prompt Installation

Copy this prompt to OpenClaw to install it automatically.

Help me install langsearch using Clawhub. If Clawhub is not installed, install it first (npm i -g clawhub).

Prefer to download?

Get the raw skill files in a ZIP archive.

What is LangSearch?

LangSearch provides a robust interface for developers to equip their agents with real-time internet access and high-precision data retrieval. By combining traditional keyword search with vector-based semantic matching, it allows Openclaw Skills to pull relevant information from billions of web documents. This skill is essential for overcoming LLM knowledge cutoff dates and ensuring that generated responses are grounded in current, factual data.

Beyond simple retrieval, the skill includes a specialized reranking engine that optimizes search results for maximum relevance. Whether you are building complex research agents or simple information bots, incorporating this functionality into your Openclaw Skills workflow ensures a higher level of accuracy and context-awareness in every interaction.

LangSearch Use Cases

  • Retrieving the latest technical documentation or software release notes during a coding session.
  • Automating market research by gathering and summarizing news from specific timeframes.
  • Enhancing RAG pipelines with semantic reranking to provide the most relevant context to an LLM.
  • Conducting multi-query deep dives to synthesize information from various web sources.
  • Monitoring specific topics using freshness filters to track daily or weekly developments.

How LangSearch Works

  1. The agent triggers a request to the web search endpoint using a natural language query or specific keywords.
  2. LangSearch executes a hybrid search across its index, utilizing both keyword and vector matching to find the best candidates.
  3. The API returns a list of results including titles, URLs, snippets, and optional markdown-formatted summaries.
  4. For advanced accuracy, the agent sends these results to the reranking endpoint along with the original query.
  5. The semantic reranker reorders the documents based on deep contextual relevance, providing an optimized set of data for the agent to process within the Openclaw Skills environment.

LangSearch Setup

To get started with this skill, you need a valid API key from the provider.

# 1. Sign up for a free API key at https://langsearch.com/api-keys
# 2. Set the environment variable in your terminal
export LANGSEARCH_API_KEY="your_api_key_here"

# 3. For permanent configuration, add the key to your .env file
echo "LANGSEARCH_API_KEY=your_api_key_here" >> .env

LangSearch Data Schema & Taxonomy

The skill returns structured data designed for easy consumption by AI agents and RAG systems:

Property Type Description
title string The title of the discovered web page.
url string The direct URL to the source content.
snippet string A short text excerpt for quick context.
summary string A detailed markdown summary (if requested).
score float The relevance score assigned by the search or rerank engine.

LangSearch Advanced Features

  • Freshness filtering (day, week, month, year) to ensure Openclaw Skills access the most recent data.
  • Support for the langsearch-rerank-1 model to dramatically improve retrieval precision.
  • Hybrid search architecture combining keyword and vector retrieval methods.
  • Built-in markdown summary generation for immediate context injection into LLM prompts.
  • High-limit search capabilities returning up to 100 results per query for deep-data mining.

SKILL.md


Loading

Related Openclaw Skills

METADATA

Github Stars: 0
forks: 0

Featured*