3073 Model Context Protocol servers — searchable, rated, ready to install.
Find ready-made build kits for Claude Code and Codex: slides, LINE stickers, pixel art, games.
Load planning for containers, trucks & pallets in Claude & ChatGPT, with a 3D load plan in the chat.
Concert booking intelligence: run and read Promoter Briefs from Claude or ChatGPT.
Memory database for LLM agents — persistent semantic + keyword memory, project namespacing, served o
Korean cloud for AI agents — create servers, deploy Docker apps, manage domains, call 300+ AI models
Lets one coding agent talk to another: Claude Code, Codex or Grok
Self-hosted shared memory for a team of AI agents. Rooms, L0-L3 depth, no LLM on the read path.
Browse Gopher and Gemini resources safely: SSRF protection, TLS/TOFU, structured JSON output.
Local 2B and 0.8B decision models: calibrated yes/no, choice and score answers for your agent.
Ask Claude about your meetings: summaries, action items and transcripts from Roomtone
Build an agent tool once; ship it as Agent Skills, MCP servers, Agent Plugins / Claude / Codex
Run Claude Code subagents on cheaper opencode models (DeepSeek, opencode Go) with native parity.
Upload files and get instant public CDN URLs from Claude Desktop, Cursor, and other MCP hosts.
Pronunciation assessment, phoneme scoring, speaker voice ID, audio transcription, speech synthesis.
Send a fax from Claude, ChatGPT, or Cursor. Upload a PDF, pay, get delivery status.
Ocean acoustics: propagation models, bathymetry, sound speed, vessel noise, live AIS. Open.
Pre-injection context compression for coding agents. Zero LLM calls, zero telemetry, offline-safe.
Leanroute MCP: route to 13 LLM providers, check pricing/usage/credit, manage keys from any agent.
Guardian agent for AI coding: four frontier models review risky diffs and commits before they ship.
Commission a multi-model AI spec committee from your agent; get rubric-scored, build-ready specs.
Ask a company's DeskClone support agent from Claude, Cursor or any MCP client (API key needed).
Instruments for what a model cannot know on its own: time, randomness, DNS, hashes, receipts.
Cited EOR provider fees, coverage and models, plus employer costs in 139 countries.
Model and map any 2028 U.S. race down to the precinct: scenarios, swing, turnout, and roll-ups.
POST /mcp