agent-search-mcp
★ 112
vs
wet
★ 19
agent-search-mcp vs wet
agent-search-mcp: Agent Search MCP is an open-source, self-hosted web search router designed for AI agents, providing free, multi-source evidence with robust cost control. It supports both English and Chinese queries, offering compact output to reduce token costs for agent interactions.; wet: WET is an open-source MCP server designed to empower AI agents with advanced web capabilities. It provides web search, content extraction, and library documentation indexing, offering comprehensive data retrieval for agent-driven tasks.
01
TL;DR
agent-search-mcpRAG / Knowledge Base · ★ 112
Integrating robust web search into AI agent clients such as Claude Desktop, Cursor, or VS Code
wetRAG / Knowledge Base · ★ 19
AI Agent Web Exploration: Empower AI agents to conduct comprehensive web searches and retrieve relevant content for diverse tasks, facilitating informed decision-making.
02
Side-by-Side Comparison
Field
agent-search-mcpwetCategory
RAG / Knowledge Base
RAG / Knowledge Base
Stars
★ 112
★ 19
License
Apache-2.0
Apache-2.0
Install Command
npx -y agent-search-mcp
pip install wet
Updated
1mo ago
1d ago
Open Source
Yes
Yes
Tags
Web Search, AI Agents, Search Router
AI Agents, Web Search, Content Extraction
03
Features
agent-search-mcp01Free web search with zero-key sources, eliminating the need for an API account
02Control provider costs with explicit routing policies for paid services
03Reduce token costs for LLMs through compact output and evidence budgeting
04Aggregates multi-source evidence, preserving provenance and handling partial failures
05Supports Chinese web search via Sogou and Baidu without a translation layer
wet01Web Search: Embedded SearXNG metasearch (Google, Bing, DuckDuckGo, Brave) with query expansion, TTL cache, standardized citation format, and 200-token snippet cap. Optional cloud search backends as a fallback chain.
02Academic Research: Search across major academic databases including Google Scholar, Semantic Scholar, arXiv, PubMed, CrossRef, and BASE.
03Library Docs: Automatically discover and index library documentation using FTS5 hybrid search, HyDE-enhanced retrieval, and support for version-specific queries.
04Content Extraction: Utilizes a 5-strategy escalation chain for robust URL content extraction, producing smart chunks with structured output (text, markdown, JSON-LD, code blocks, metadata), supports batch processing, deep crawling, local file conversion, and includes anti-bot strategies.
05Zero Config Fallbacks: Offers out-of-the-box functionality with built-in local reference embedding and reranking via fastretrieval, requiring no API keys, while also supporting optional cloud providers for enhanced quality.
04
Use Cases
agent-search-mcp↳Integrating robust web search into AI agent clients such as Claude Desktop, Cursor, or VS Code
↳Enabling AI agents to gather factual information and verify claims from multiple web sources
↳Developing agent-authored answers from cited evidence while managing token budgets effectively
↳Performing multilingual web searches, particularly for Chinese content, for various AI applications
wet↳AI Agent Web Exploration: Empower AI agents to conduct comprehensive web searches and retrieve relevant content for diverse tasks, facilitating informed decision-making.
↳Automated Research & Knowledge Gathering: Automate academic and library documentation research, efficiently gathering and providing structured information to AI agents for analysis and synthesis.
↳Content Analysis & Summarization: Facilitate advanced content extraction from URLs and local files, preparing data for in-depth analysis, summarization, or integration into Retrieval-Augmented Generation (RAG) systems.
↳Developer Documentation Lookup: Grant AI agents version-aware access to comprehensive library and project documentation, aiding in coding assistance, debugging, and issue resolution.
FAQ
FAQ
What is the difference between agent-search-mcp and wet?
Both agent-search-mcp and wet are in the RAG / Knowledge Base category. agent-search-mcp has 112 stars, while wet has 19 stars.
Which is better, agent-search-mcp or wet?
The best choice depends on your use case. Choose agent-search-mcp if Integrating robust web search into AI agent clients such as Claude Desktop, Cursor, or VS Code, and wet if AI Agent Web Exploration: Empower AI agents to conduct comprehensive web searches and retrieve relevant content for diverse tasks, facilitating informed decision-making..
Is agent-search-mcp free or open source?
Yes, agent-search-mcp is open source on GitHub (Apache-2.0).
Is wet free or open source?
Yes, wet is open source on GitHub (Apache-2.0).
Cloud Infrastructure
AdDeploy on DigitalOcean
Host your MCP servers, Docker containers, and AI agents with $200 free credit over 60 days.
→