RyRank

Name: RyRank
Author: trigeochiral

Local

trigeochiral.comSearch0Updated 8w ago

Repository Visit website

About RyRank

RyRank is an MCP server that reranks documents by relevance to a query, aimed at RAG and document or tool shortlists. It returns Cohere-compatible results and also supports the Cohere v2 and Voyage rerank API shapes.

Key features

01Query-based document reranking
02Cohere-compatible response format
03Cohere v2 and Voyage API shapes
04Candidate narrowing, deduplication, and triage

Use cases

01Rerank documents retrieved for RAG
02Prioritize a shortlist of tools or documents
03Deduplicate and triage retrieval candidates

MCP support

Exposes a `rerank` tool to MCP clients for ranking documents against a query.

Configuration

Local server config is the same for every client. Copy it into your client’s MCP config file.

mcp.json · Universal config

{
  "mcpServers": {
    "trigeochiral-ryrank": {
      "command": "uvx",
      "args": [
        "ryrank-mcp"
      ],
      "env": {
        "RYRANK_API": "<RYRANK_API>",
        "RYRANK_XPAYMENT": "<RYRANK_XPAYMENT>"
      }
    }
  }
}

Run commanduvx ryrank-mcp

Environment variables to set

RYRANK_APIBase URL of the hosted RyRank API (default https://trigeochiral.com)

RYRANK_XPAYMENTSecretOptional base64 x402 X-PAYMENT payload for paid calls past the free tier

Local servers need a runtime on your machine (e.g. Node.js 18+). Config file locations differ by client; see your client’s docs.

Packages

pypiryrank-mcp1.0.0stdio

Tools

Saved tool list · updated 2w ago

Parameter	Type	Required	Description
query	string	Yes	the query to rank against
top_n	integer	No	return only the top N (default: all)
documents	array	Yes	2-1000 documents (strings)
return_documents	boolean	No	include the document text in each result

Details

From the project README

RyRank — MCP server

A Cohere-compatible reranker, pay-per-call in USDC on Base (x402), free tier, no signup.

Hosted API: https://trigeochiral.com — set it as RYRANK_API.

Tool

Tool	What
`rerank`	Rerank documents by relevance to a query. Cohere-compatible response shape (`results: [{index, relevance_score}]`), so it drops into an existing `co.rerank()` call. ~$0.001/call over x402 (roughly half Cohere Rerank's list price) or a free tier.

Why

A hybrid reranker, not a cross-encoder: strong when the candidates already share vocabulary with the query, which is the common case for RAG retrieval output and document/tool shortlists. Built for high-volume, latency- and cost-sensitive reranking — narrowing a retrieval candidate set, deduping, triage — rather than close semantic disambiguation between near-identical phrasings. Pay-per-call and about half the price is the trade for that.

Verified against the real cohere.Client, cohere.ClientV2 and voyageai.Client SDKs: point base_url at this API and existing rerank code runs unmodified.

Install

pip install ryrank-mcp
claude mcp add ryrank --env RYRANK_API=https://trigeochiral.com -- ryrank-mcp

or with uvx (no install):

claude mcp add ryrank --env RYRANK_API=https://trigeochiral.com -- uvx ryrank-mcp

Optional: RYRANK_XPAYMENT = a base64 x402 X-PAYMENT payload, to make paid calls past the free tier. Without it you get 25 free calls/day per IP.

Direct HTTP (no MCP)

curl -sX POST https://trigeochiral.com/v1/rerank -H 'content-type: application/json' \
  -d '{"query":"how do I reset my password","documents":["Go to Settings > Security and click Reset Password.","Our refund policy allows returns within 30 days."],"top_n":2}'

Also speaks the exact Cohere v2 (POST /v2/rerank) and Voyage (POST /voyage/v1/rerank) request and response shapes, so their official SDKs work unmodified with just a base_url override — see https://trigeochiral.com for details.

License

MIT

Comments

Markdown supported

New to MCP?

From what the protocol is to installing your first server, these answers cover most first questions.

Read the getting-started guide

An open protocol that defines how AI applications connect to external tools and data in a uniform way. Clients talk to the model; MCP servers provide tools, resources and prompt templates. They exchange standard messages, so one server can be reused by many clients.