Cheaprouter MCP
Routes each LLM completion to the cheapest capable provider across 8 providers (BYOK).
Model inference and embedding endpoints, including routers that front several providers. Context limits and prices change often, so treat the price with its check date.
Routes each LLM completion to the cheapest capable provider across 8 providers (BYOK).
Read-only search of your Sortio knowledge graph (files and entities) for Claude and ChatGPT.
Read-only-by-default MCP server: let LLMs explore the Fediverse — Mastodon, Misskey, Pleroma.
MCP server for querying Foxhound traces from Claude Code, Cursor, and other MCP clients
Deterministic SVG perception for LLM agents — measures, never guesses. MCP server.
A second opinion for AI agents: one prompt across several live Gonka models + roles, one call.
Stop copy-pasting between Claude Chat and Claude Code.
Brazilian agricultural data for LLMs — prices, crops, climate, deforestation.
Social layer for Claude Code - DMs, presence, games between AI-assisted developers
Catch bad animations before they ship. Deterministic motion audit + vision-LLM design review.
Control FreeCAD from Claude — 3D modeling, PartDesign, CAM toolpaths, and more via 30+ tools.
Model inventory and governance: trace deployed models and their dependencies as an append-only log
A content exploration toolkit that helps LLMs surface high signal, unsummarized web content.
Tells you how you and your Claude agent actually work together. Local-only, no API keys.
Which parts of an ML need Hugging Face already covers. Every model and dataset checked live.
MCP protocol enforcing explicit task acceptance and attestation-tagged completion claims.
Filesystem-first MCP for Obsidian — an LLM-maintained wiki inspired by Karpathy's LLM Wiki.
Design systems, UI prompts, and layout variations for AI coding tools.
Open-source CRM: Claude, ChatGPT or any MCP client works with records, inbox and company knowledge.
Read and sign the guestbook of Morphogen II, a living reaction-diffusion field with open questions.
Let Claude use your Zen Browser, like Claude in Chrome. No extension, only sites you allow.
Continuous behavioral drift monitoring for LLM apps — catches silent provider model updates.
Prompt-injection firewall for AI agents — scan untrusted text before LLM calls.
Run commands with API keys injected per call — values never enter model context.