cratesio-mcp

Gives your AI agent crate search, documentation, dependency analysis, download stats, and security auditing to make informed decisions about Rust dependencies.

{ "mcpServers": { "cratesio-mcp": { "type": "http", "url": "https://cratesio-mcp.fly.dev/" } } }

git clone https://github.com/joshrotenberg/cratesio-mcp cd cratesio-mcp cargo install --path .

docker run -p 3000:3000 ghcr.io/joshrotenberg/cratesio-mcp:latest

{ "mcpServers": { "cratesio-mcp": { "command": "cratesio-mcp" } } }

{ "mcpServers": { "cratesio-mcp": { "type": "http", "url": "http://localhost:3000/" } } }

{ "mcpServers": { "cratesio-mcp": { "command": "cratesio-mcp" } } }

Every successful tool call returns both the existing human-readable Markdown incontentand typed JSON instructuredContent. Each tool advertises the matching JSON Schema 2020-12 contract throughoutputSchemaduring discovery.

- stdio(default): for Claude Desktop, Claude Code, and other MCP clients
- Streamable HTTP: sessionless MCP 2026-07-28 transport, while retaining compatibility with session-based MCP 2025-11-25 clients

# stdio (default) cratesio-mcp # HTTP cratesio-mcp --transport http --port 3000

The 2026-07-28 path supportsserver/discover, stateless requests, standardized MCP routing headers, and cache hints. Tool discovery marks every crates.io tool as read-only, non-destructive, and idempotent; resource reads advertise cache lifetimes appropriate to their data source.

cratesio-mcpis built on its own typed async crates.io API client, withnocrates_io_apidependency. You can use it directly as a library:

- ~46 endpointsacross crates, versions, owners, categories, keywords, users, teams, API tokens, publishing, and trusted publishing.
- Full read and write coverage:search and metadata, plus authenticated operations (publish, yank/unyank, add/remove owners, manage API tokens, configure trusted publishing) via.with_auth(token).
- Resilient by default:built-in rate limiting (respects the crates.io crawling policy) and retry with exponential backoff on transient failures (429 / 5xx).
- Documented and tested:every public method has doc comments, with a wiremock test suite covering the endpoints (including the authenticated write paths).

use std::time::Duration; use cratesio_mcp::client::{CratesIoClient, CratesQuery, Sort}; let client = CratesIoClient::new("my-app", Duration::from_secs(1))?; // Search for crates let query = CratesQuery::builder() .search("tower") .sort(Sort::Downloads) .build(); let results = client.crates(query).await?; // Get crate details let info = client.get_crate("tower-mcp").await?;

The OSV.dev vulnerability client (audit_dependencies) and the docs.rs rustdoc-JSON client (get_crate_docs/get_doc_item/search_docs) ship alongside it.

This is a web browser that enables your coding agent, such as Claude Code, to visit websites on your behalf and assist you in identifying bugs or creating UI test cases.

Create crafted UI components inspired by the best 21st.dev design engineers.

Bring agent evaluations, observability, and synthetic test set generation directly into your IDE for free with Galileo's new MCP server

An MCP server to help AI assistants to answer questions and generate AccelByte Extend SDK code more effectively .

MCP server for AI Diagram Maker — generate beautiful software engineering diagrams directly inside Cursor, Claude Desktop, Claude Code, or any MCP-compatible AI agent

ALAPI MCP Tools,Call hundreds of API interfaces via MCP

AI-powered SVG animation generator that transforms static files into animated SVG components using the Allyson platform

MCP server that gives AI assistants on-demand access to 1,500+ amCharts docs, ~300 code examples, and 1000+ class API references.

APIMatic MCP Server is used to validate OpenAPI specifications using APIMatic. The server processes OpenAPI files and returns validation summaries by leveraging APIMatic’s API.

One shared context layer for AI agents and humans — live API specs, DB schemas, and versioned contracts across repos so every agent and teammate works from the same source of truth.