yandex-mcp
Enables LLM agents to interact with Yandex Tracker (and future Yandex services) using per-user credentials instead of a shared org-wide token, supporting both single-user and multi-user modes.
README
Archived — superseded by aikts/yandex-tracker-mcp
This repository is archived. Do not use it.
After reading the code of existing Yandex Tracker MCP servers (not just their READMEs), we found aikts/yandex-tracker-mcp is a mature superset of everything built here:
- Standalone per-user OAuth profile — the same MCP-SDK
OAuthAuthorizationServerProviderarchitecture (authorize → Yandex, callback, server-side code exchange, no token passthrough, refresh+rotation+revoke, DCR, RFC 8707), with redis/memory/encrypted stores. (Ourpackages/tracker-mcp/src/auth/*was a less-mature duplicate.) - Gateway / reverse-proxy Bearer passthrough — reads a per-user Yandex token
from
Authorization: Bearerfor deployments where a gateway (e.g. Faust) holds the tokens. Same as our v0.1 HTTP bearer mode. - Plus IAM tokens, service-account auth, both
X-Org-ID/X-Cloud-Org-ID, caching, a wide tool surface, elicitation confirmations. - Apache-2.0, Python (Python is also the language of our agent runtime, Gena).
Per the decision gate (reuse established OSS rather than maintain a duplicate), we adopt aikts and retire this repo.
What carries over (verified here, still valid for aikts)
These facts were verified while building this repo and apply equally to aikts / Tracker v3:
- Tracker OAuth scopes:
cloud:auth+tracker:read+tracker:write(cloud:authis the one that unblocks Cloud-org access; without it Tracker returns 401 even with the tracker scopes). - Cloud-org auth:
Authorization: OAuth <token>+X-Cloud-Org-Id: <org-id>— direct OAuth, no IAM exchange (the OAuth→IAM exchange is dead for federated Cloud users on tokens issued after 2026-06-01). - Tracker v3 search body is top-level
{query};perPage/pageare query params.
Reusable dev helpers (not server-specific)
These three scripts in this repo work for any Yandex Tracker OAuth setup:
scripts/fetch-token.ts— local OAuth code-flow helper to obtain a Tracker access token (run the dance once).scripts/discover-org.ts— non-admin Cloud-org id discovery via IAM.packages/tracker-mcp/scripts/live-probe.ts— spawn any stdio Tracker MCP and callget_myself/search_issueslive to verify it works.
Keep them as references; they are not published as a package.
Recommended Servers
playwright-mcp
A Model Context Protocol server that enables LLMs to interact with web pages through structured accessibility snapshots without requiring vision models or screenshots.
Magic Component Platform (MCP)
An AI-powered tool that generates modern UI components from natural language descriptions, integrating with popular IDEs to streamline UI development workflow.
Audiense Insights MCP Server
Enables interaction with Audiense Insights accounts via the Model Context Protocol, facilitating the extraction and analysis of marketing insights and audience data including demographics, behavior, and influencer engagement.
VeyraX MCP
Single MCP tool to connect all your favorite tools: Gmail, Calendar and 40 more.
graphlit-mcp-server
The Model Context Protocol (MCP) Server enables integration between MCP clients and the Graphlit service. Ingest anything from Slack to Gmail to podcast feeds, in addition to web crawling, into a Graphlit project - and then retrieve relevant contents from the MCP client.
Kagi MCP Server
An MCP server that integrates Kagi search capabilities with Claude AI, enabling Claude to perform real-time web searches when answering questions that require up-to-date information.
E2B
Using MCP to run code via e2b.
Neon Database
MCP server for interacting with Neon Management API and databases
Exa Search
A Model Context Protocol (MCP) server lets AI assistants like Claude use the Exa AI Search API for web searches. This setup allows AI models to get real-time web information in a safe and controlled way.
Qdrant Server
This repository is an example of how to create a MCP server for Qdrant, a vector search engine.