MCP Servers

Deterministic documentation-drift detection for AI agents. Validates a project's canonical docs against its actual code — no LLM guessing; every finding has a stable code and a suggested fix. Five read-only tools over stdio: guard, score, explain, verify-claims, diagnose. npx -y docguard-cli mcp

EOQ-based context economics for coding agents. Monitors LLM token cache with inventory theory, signals optimal session restart timing via live dashboard and terminal statusline. Zero context pollution — no metric ever enters the model's context.

A semantic search engine for files and code — find files by describing what they do, not their names. Local reusable indexes, configurable embedding providers (OpenAI, Gemini, Voyage, any OpenAI-compatible API, or fully offline local models), AST-aware code chunking, optional reranking. uvx vexor mcp or pip install vexor.

Check whether an LLM model id is deprecated, retiring, or retired, and what to migrate to. Covers OpenAI, Anthropic, Azure, Bedrock, Google, and Cohere. No API key. npx -y @llmintel/mcp or https://llmintel.ai/v1/mcp

🎖️ - Connector between LLM agents and world data - 1,500+ endpoints aggregating 160+ primary sources across 36 data domains: markets, macroeconomics, company fundamentals, government, news, climate, maritime, and entity screening. Install via pip install sugra-api-mcp or hosted at app.sugra.ai/mcp.

Convert unstructured business documents (PDF, DOCX, invoices, contracts, payroll) into structured, LLM-ready datasets with automatic classification, field extraction, PII masking (10+ locales), and quality scoring. 6 async MCP tools covering the full pipeline: submit → poll → extract → build → export JSONL/RAG/CSV. Install: pip install flexorch-mcp.

notmuch MCP server. Wraps the notmuch CLI to give your LLM access to a local Maildir: full-text search, tag filtering, thread view, attachment reading (PDF, office docs, .ics), related-thread lookup, and unanswered-mail tracking. Optionally write plain-text drafts into a local maildir (--allow-drafts) or manage tags (--allow-tags). No SMTP, no sendmail.

Convert .eml email exports to clean Markdown with YAML front matter — thread splitting, signature stripping, attachment extraction, and calendar parsing. Tuned for feeding archived email into RAG/LLM pipelines.

Full SSH access for AI agents with secret-safe environment injection and automatic output redaction. Persistent sessions, structured execution, SFTP, port forwarding — secrets are fed via stdin (invisible to ps) and scrubbed from all tool responses before reaching the LLM. uvx mcp-remote-ssh

MCP server for VibeKit. Lets an LLM in Claude Desktop, Cursor, or any MCP client build, deploy, and manage hosted apps and chat with each app's persistent server-side agent (every app gets its own agent and a live URL). Not to be confused with the unrelated superagent-ai/vibekit SDK.

Run GLM (Zhipu/Z.ai) as a real sub-agent inside Claude Code or GitHub Copilot. GLM gets its own agent loop (read/write/edit/run) on your repo — not a single LLM call — with peak-aware Opus-vs-GLM routing, diff/dry-run/git-revert oversight, and a usage ledger. ~10x cheaper than Opus. Requires a Z.ai GLM Coding Plan key.

A council of 49 specialist agents + 93 tools for every major AI CLI (Claude Code, Codex, Gemini, Cursor, Windsurf). Deterministic agents that optionally upgrade to LLM, a self-learning router, cross-CLI memory, and guards for dependency hallucinations and decision drift. No API keys; zero extra cost on subscriptions.

Multi-agent coordination bus over MCP: atomic gap-free claims, advisory file leases, a shared event ledger, presence, prose handoffs, and a task graph, so multiple coding agents (Claude Code, Cursor, Codex) stop colliding and re-deriving each other's work. Remote Streamable HTTP; self-hostable (AGPL-3.0) or hosted at llm-bus.com.

️ - Build and run a complete backend from your agent on the Anythink platform: relational data with row-/field-level security, full-text + semantic + geo search, RBAC + BYOK, a workflow/automation engine, a growth & retention engine (email, actionable push, promotions, per-user referral codes + rewards, points/credits), payments + marketplace billing, and a growing catalog of integrations (Claude, OpenAI, Slack, Google, GitHub…) — driven through one CLI-backed MCP tool. Install with npx -y @anyt

Pipe one-click browser-tab screenshots into any MCP client, 100% local — a companion extension captures, the local server stacks them and serves them to your LLM on demand.

"Mermaid for human movement" — a text language any LLM can write, rendered as a safe, anatomically-clamped 3D human figure in the browser.

Catch bad animations before they ship. 13 deterministic checks on the live page (duration, easing, stagger, exits, reduced-motion, touch-gated hover) with no API key or LLM, plus optional vision-LLM UX review across viewports. Install via NPM: npx -y motionlint mcp

A ready-to-clone Rust MCP server that calls Anthropic's Claude API via the crimson-crab SDK. Exposes an askclaude tool. MIT/Apache-2.0.

TokenLab AI gateway MCP server for model and pricing discovery, OpenAI-compatible Chat Completions, and native Responses, Anthropic Messages, and Gemini inference. Catalog tools require no API key; inference tools use an optional TokenLab key.

Daily-verified LLM API pricing dataset (44+ models, CN & global) with a hosted MCP server for live price queries and token cost estimation.

Pay-per-call x402 gateway: one MCP server for 90+ agent tools (utilities, DNS/WHOIS, blockchain RPC, market data, prediction markets, DEX data, security audits) plus an OpenAI-compatible LLM gateway. USDC on Base, free-trial calls per agent. pip install ddg-agent-services-mcp or remote https://mcp.daedalusdevelopmentgroup.com/mcp.

Contract validation and self-healing failover for LLM APIs. 6-dimension verification (structure, schema, latency, cost, identity, integrity) in 22μs P50. 87 self-healing rules with MAPE-K autonomic loop. BYOK direct connect to 9 providers (OpenAI, Anthropic, DeepSeek, Moonshot, Zhipu AI, Qwen, SiliconFlow, Groq, Together AI). L3 failover in 949ms E2E. Install: npx -y correctover-mcp-server.

Open-source LLM knowledge base for teams. 28-tool native MCP server turns Slack/Discord/Teams/Mattermost chat into a typed knowledge graph + auto-generated wiki with cited answers, semantic search, expert finding, and decision tracing. BYO LLM via LiteLLM, Apache 2.0, on-prem via Docker.

Temporal memory MCP server with LanceDB vector search, weight-decay scoring, and LLM consolidation. Automatic session capture, deduplication, 6-factor relevance scoring. pip install mem-context