Valyu

Valyu is an MCP server that integrates Valyu's knowledge retrieval and feedback APIs into Model Context Protocol environments. It connects language models directly to proprietary indexes and web sources, enabling AI agents to pull relevant context into conversations and workflows. Developers, researchers, and automated assistant builders use this server to retrieve grounded factual data from specific indexed sources such as arXiv or Wikipedia. Beyond search capabilities, the server supports submitting transaction-level feedback, allowing systems to log user sentiments and quality assessments. Queries can be fine-tuned by defining source types, maximum query price thresholds, result counts, and minimum similarity scores. Query rewriting is also supported to optimize search quality. By providing standardized tool calls for both information retrieval and transaction feedback, Valyu functions as a context engine and feedback collector for conversational interfaces and autonomous agents.

Category: AI Memory & Context

Tags: feedback, information-retrieval, knowledge-retrieval, rag

Visit Valyu

How to install and configure Valyu

1. Pull the Docker image by running: docker pull ghcr.io/tiovikram/valyu-mcp-server 2. Open your Claude Desktop configuration file and add the server definition under mcpServers: json "mcpServers": { "valyu": { "command": "docker", "args": ["run", "--pull", "--rm", "-i", "-e", "VALYU_API_KEY", "ghcr.io/tiovikram/valyu-mcp-server"], "env": { "VALYU_API_KEY": "<your-valyu-api-key>" } } } 3. Replace <your-valyu-api-key> with your valid Valyu API key and restart your MCP client.

What you can do with Valyu

• Querying proprietary indices and open web documentation to ground language model answers in verified source material. - Setting search parameters such as cost ceilings, similarity thresholds, and maximum result limits to control retrieval expenses. - Submitting structured user sentiment and quality feedback linked to specific transaction identifiers for auditing and model evaluation. - Automatically rewriting search prompts within agent workflows to enhance information retrieval accuracy across diverse data sources.

Key facts

• https://github.com/valyu-network/valyu-mcp-js
• AI Memory & Context, Web Search & Research
• feedback, information-retrieval, knowledge-retrieval, rag

Part of MCP Servers

How do I install Valyu?

You can deploy Valyu using Docker. Pull the container image ghcr.io/tiovikram/valyu-mcp-server, then add the Docker execution command to your client settings file alongside your Valyu API key environment variable.

What can Valyu do?

Valyu provides two primary tools: knowledge, which searches web and proprietary data sources with customizable filters and price limits, and feedback, which records user sentiment and transaction notes.

Which MCP clients work with Valyu?

Valyu works with any Model Context Protocol compliant client that supports Docker-based command execution, including Claude Desktop, Cursor, and the official MCP inspector tool.

What parameters are required to run a knowledge search in Valyu?

A knowledge search requires a query string, a search_type specifying proprietary, web, or all, and a max_price representing the maximum cost per thousand queries.