Model access, evaluation and agent infrastructure.
Cadrage prêt à l'emploi pour développer une application avec un agent de code (Claude Code, Codex, Cursor…) : PRD, architecture, specs, ADR, skills, hooks, CI.
MCP server for HTML generation from AI agents
🔬 Local SQLite-based AI research agent swarm with GOAP planning, multi-perspective analysis, long-horizon recursive framework, AgentDB self-learning, anti-hallucination controls, and MCP server. Goal-oriented with parallel execution. No cloud dependencie
AA Claw 🦀 — Lightweight AI Agent gateway in Rust. Single static binary, any OpenAI-compatible provider, built-in memory, remote control & multi-agent orchestration. Auth via aaclaw.net.
MCP server for fetching actual free models from OpenRouter API
Local RAG MCP Server - Easy-to-setup document search with minimal configuration
SuperBryn MCP server — run the voice-agent eval loop and read your production call logs from Claude Code, Claude Desktop, or Codex.
MCP server and CLI for image generation/editing via OpenRouter's Responses API
MCP server exposing the pi coding agent as a delegable, steerable worker
Scaffold a production-ready Express 5 + TypeScript backend — pick a database + ORM (Postgres/SQLite/Drizzle or Mongo/Mongoose), auth (Auth0/JWT/Clerk), an optional Next.js front end, Sentry, an MCP server, and x402 payments, and get a project with only wh
379 AI & Dev skills for Claude Code, Cursor, Windsurf, VS Code, and JetBrains — install via MCP.
SpiderIQ Publish: content platform MCP (pages, posts, templates, components, domains, media)
Controlled A/B testing for AI agent skills, .cursorrules, CLAUDE.md, and system prompts: blind grading, bootstrap confidence intervals, and HELPS/PLACEBO/HARMS effect size measurement.
The official ARD Registry CLI and client library. Search and navigate MCP servers, A2A agents, and AI skills by intent, at runtime.
Discover monetized AI tools on SettleGrid. Search by category, get pricing.
Local-first intelligent MCP router: one endpoint, many MCP servers, capabilities discovered on demand.
prosemeter MCP server — deterministic prose scoring and the agent revise/converge loop over MCP
AI-agent-friendly command-line toolkit: timestamp/date conversion, JWT decoding, port killing, file finding, and directory trees — headless, --json, with a self-describing manifest and a built-in MCP server (mfn mcp).
Multi-agent LLM platform with a visual workflow engine — chat with agents directly or build deterministic node-graph pipelines they execute step by step.
Local AI agent API integrations for GitHub, Notion, Slack, Linear, and Stripe with JSON manifests, approval checks, and vault keys.