AI Skill Hub 推荐使用:智能搜索工具 是一款优质的MCP工具。AI 综合评分 7.5 分,在同类工具中表现稳健。如果你正在寻找可靠的MCP工具解决方案,这是一个值得深入了解的选择。
智能搜索工具 是一款遵循 MCP(Model Context Protocol)标准协议的 AI 工具扩展。通过 MCP 协议,它可以让 Claude、Cursor 等主流 AI 客户端直接访问和操作外部工具、数据源和服务,实现 AI 能力的无缝扩展。无论是文件操作、数据库查询还是 API 调用,都可以通过自然语言在 AI 对话中直接触发,极大提升生产效率。
智能搜索工具 是一款遵循 MCP(Model Context Protocol)标准协议的 AI 工具扩展。通过 MCP 协议,它可以让 Claude、Cursor 等主流 AI 客户端直接访问和操作外部工具、数据源和服务,实现 AI 能力的无缝扩展。无论是文件操作、数据库查询还是 API 调用,都可以通过自然语言在 AI 对话中直接触发,极大提升生产效率。
# 方式一:通过 Claude Code CLI 一键安装
claude skill install https://github.com/brcrusoe72/agent-search
# 方式二:手动配置 claude_desktop_config.json
{
"mcpServers": {
"------": {
"command": "npx",
"args": ["-y", "agent-search"]
}
}
}
# 配置文件位置
# macOS: ~/Library/Application Support/Claude/claude_desktop_config.json
# Windows: %APPDATA%/Claude/claude_desktop_config.json
# 安装后在 Claude 对话中直接使用 # 示例: 用户: 请帮我用 智能搜索工具 执行以下任务... Claude: [自动调用 智能搜索工具 MCP 工具处理请求] # 查看可用工具列表 # 在 Claude 中输入:"列出所有可用的 MCP 工具"
// claude_desktop_config.json 配置示例
{
"mcpServers": {
"______": {
"command": "npx",
"args": ["-y", "agent-search"],
"env": {
// "API_KEY": "your-api-key-here"
}
}
}
}
// 保存后重启 Claude Desktop 生效
Self-hosted search API for AI agents. 16 endpoints. 9-strategy content extraction. Optional Tor-anonymized stack. No third-party search API keys, no per-query fees, no vendor lock-in. Optional local bearer auth is supported.
git clone https://github.com/brcrusoe72/agent-search.git
cd agent-search
./scripts/prepare-searxng.sh
docker compose up -d
curl "http://localhost:3939/search?q=distributed+consensus+algorithms"
You now have a deduplicated, multi-engine search API running on :3939.
If you enable auth, pass the token on all non-health endpoints:
export AGENT_SEARCH_TOKEN="change-me"
curl -H "Authorization: Bearer $AGENT_SEARCH_TOKEN" \
"http://localhost:3939/search?q=distributed+consensus+algorithms"
Prefer not to use Docker for the API server?
git clone https://github.com/brcrusoe72/agent-search.git
cd agent-search
./scripts/install-native.sh
./scripts/run-native.sh
Native mode requires Python 3.11+ and a reachable SearXNG instance with JSON output enabled. It stores AgentSearch state in ./data. See Native Install.
Environment variables (set in docker-compose.yml or .env):
| Variable | Default | Description |
|---|---|---|
SEARXNG_URL | http://searxng:8080 | SearXNG instance URL |
SEARXNG_IMAGE | pinned SearXNG digest | SearXNG container image; override only when intentionally upgrading |
PYTHON_BASE_IMAGE | pinned Python digest | API Docker base image; override only when intentionally upgrading |
COREDNS_IMAGE | pinned CoreDNS digest | Private-stack DNS image |
SOCAT_IMAGE | pinned socat digest | Private-stack TCP forwarder image |
TOR_BASE_IMAGE | pinned Debian digest | Private-stack Tor proxy base image |
CACHE_TTL | 3600 | Cache duration in seconds |
RATE_LIMIT | 60 | Max requests per minute |
SQLITE_TIMEOUT | 1.0 | SQLite lock wait timeout in seconds for query stats |
AGENT_SEARCH_TOKEN | *(empty)* | Bearer token for auth (optional) |
ADAPTERS_DIR | /app/adapters | Path to pluggable adapter modules |
pip install agentsearch-client
```python from agentsearch import AgentSearch
client = AgentSearch() # defaults to localhost:3939 results = client.search("manufacturing OEE best practices", count=5) for r in results.results: print(f"{r.title} — {r.url}")
| Module | LOC | What it does |
|---|---|---|
killchain.py | 1016 | 9-strategy escalating content extraction |
main.py | 920 | FastAPI app, 16 endpoints, auth, rate limiting |
source_tracer.py | 620 | Source provenance tracking and citation chains |
scrubber.py | 539 | Prompt injection detection and content sanitization |
source_library.py | 310 | Curated institutional source registry |
domain_trust.py | 311 | Domain trust scoring (TLD, age, reputation) |
evolver.py | 301 | Self-improvement engine — failure analysis → config tuning |
content_cache.py | 241 | URL-keyed content cache with TTL |
query_expansion.py | 201 | Server-side query variation and fusion |
Plus: 5 pluggable adapters (Cloudflare bypass, Medium, 403 handler, parse error recovery, empty content fallback), MCP server, Python SDK, test suite.
高质量的开源MCP工具,支持AI代理和自托管搜索
AI Skill Hub 为第三方内容聚合平台,本页面信息基于公开数据整理,不对工具功能和质量作任何法律背书。
建议在沙箱或测试环境中充分验证后,再部署至生产环境,并做好必要的安全评估。
✅ MIT 协议 — 最宽松的开源协议之一,可自由商用、修改、分发,仅需保留版权声明。
总体来看,智能搜索工具 是一款质量良好的MCP工具,在同类工具中具备一定竞争力。AI Skill Hub 将持续追踪其更新动态,建议收藏备用,结合自身场景选择合适时机引入使用。
| 原始名称 | agent-search |
| 原始描述 | 开源MCP工具:Self-hosted search API + MCP server for AI agents. Bundles SearXNG. Zero API key。⭐40 · Python |
| Topics | ai-agentsfastapillm-toolsmeta-search |
| GitHub | https://github.com/brcrusoe72/agent-search |
| License | MIT |
| 语言 | Python |
收录时间:2026-06-07 · 更新时间:2026-06-07 · License:MIT · AI Skill Hub 不对第三方内容的准确性作法律背书。
选择 Agent 类型,复制安装指令后粘贴到对应客户端