Home › Search & Research › wet

wet

Open-source MCP server for AI agents: web search, content extraction, and library docs.

Python CloudLocal macOSWindowsLinux

★ 18 stars Last pushed 2026-09-23 License: Apache-2.0

View repository →

Connect

Review any command before running it. Package names and URLs come from the server's own registry entry.

Package (pypi 3.15.10)

uvx wet-mcp

Or add to your MCP client config:

{
  "mcpServers": {
    "wet": {
      "command": "uvx",
      "args": [
        "wet-mcp"
      ],
      "env": {
        "API_KEYS": "<YOUR_VALUE>",
        "GITHUB_TOKEN": "<YOUR_VALUE>"
      }
    }
  }
}

API_KEYS secret — Provider API keys (format: PROVIDER_API_KEY:key,...). Select models per task with EMBEDDING_MODELS / RERANK_MODELS / LLM_MODELS (CSV provider/model, order = litellm fallback); provider is inferred from the model prefix. Empty embedding/rerank chain falls back to fastretrieval's built-in local reference model.
GITHUB_TOKEN secret — GitHub personal access token for higher rate limits on library discovery