New

MCP server exposing Foldkit DevTools to AI agents (Claude Code, Cursor, etc.)

A small, provider-neutral TypeScript toolkit for calling LLMs from backend code.

From web terminals for your coding agents to a self-running agent company: Claude Code and Codex in parallel git worktrees, a job board they pick work from, and one agent

Session for the Robota SDK: permission-gated tool execution, hooks, context tracking, compaction and persistence

Foundation of the Robota SDK: the Robota agent engine and the provider, tool, plugin and permission contracts

Programmatic SDK for building AI agents with Robota — provides InteractiveSession, createQuery(), command APIs, permissions, hooks, and context loading

Persistent session memory for AI coding agents that never leaves your machine — including the on-device model that reasons over it. Restores your prior decisions, open TO

Prism Coder — Local LLM fleet + MCP cognitive memory for AI agents. Wraps prism-mcp-server with the prism-coder brand CLI. Install once, run anywhere.

MCP (Model Context Protocol) client extension for the Pi coding agent — connect Pi to any MCP server

🏎️ The official FastMCP TypeScript library - build MCP servers and clients, fast 🏎️

Multi-account proxy for Claude Code and Codex: pools Claude Max, ChatGPT/Codex, API-key and third-party backend accounts, and rotates on quota

The self-hosted agent harness for small teams. Runs a team of AI agents across Telegram, WhatsApp, GitLab, GitHub, crons, webhooks and mesh tasks, on Claude Code, Codex,

Persistent, local-first memory for AI agents. Selective by design: filters what's worth keeping, links what's related, and says nothing when it doesn't know. SQLite + ONN

Trade through a Bonded Mandate Vault and get rated by the Bonded Arena from any TypeScript agent: a typed vault client with dry runs, Sentinel checks and a mandate builde

Managed Deep Agents — the `defineDeepAgent` authoring interface plus the CLI that compiles and deploys a code-first Deep Agent repository to a managed LangGraph runtime.

Memory compression system for Claude Code - persist context across sessions

Make AI agents aim higher without losing control: a Claude Code skill with two bars per task, a cheap gate check before expensive work, and an independent verifier that n

Amplitude AI SDK - LLM usage tracking for Amplitude Analytics

The OpenAI Agents SDK is a lightweight yet powerful framework for building multi-agent workflows. This package contains the logic for building realtime voice agents on th

Validates an immutable classifier + rules + policy + evidence bundle, settles requests with certified rules first, embeds and scores only when needed, and returns enforce

Claude Code in OpenCode — local CLI auth, Agent SDK harness, effort variants, tools/MCP, images & compact.

Markdown-vault agent memory with a derived, brain-inspired graph index (hippocampal indexing, spreading activation, bi-temporal facts).

Calibrated multi-head classifiers on text embeddings: train a calibrated logistic head per label, pick thresholds from recall or precision targets, evaluate with confiden

Exact, scikit-learn-verified solvers in TypeScript: weighted L2 logistic regression (Newton/IRLS), Platt scaling, weighted isotonic regression, plus calibration and evalu

The OpenAI Agents SDK is a lightweight yet powerful framework for building multi-agent workflows.

Elyntic self-improvement lab

The OpenAI Agents SDK is a lightweight yet powerful framework for building multi-agent workflows.

The OpenAI Agents SDK is a lightweight yet powerful framework for building multi-agent workflows.

Extensions for the OpenAI Agents SDK

A web interface for the pi coding agent: a browser chat UI you run with npx — no clone, no build.

Mount pi-outpost as a Shadow-DOM-isolated widget inside another web app.

Old abstractions from LangChain.js

Claude Code as an Agent Host Protocol backend: a plugin for the ahpd daemon and an Agent for @ahpd/sdk

Multi-provider LLM inference library for AI-powered trading strategies. Supports 10+ providers including OpenAI, Claude, DeepSeek, Grok, Mistral with unified API and auto

macOS automation MCP server with screen capture, UI interaction, and AI analysis

Model Context Protocol server for backtest-kit: lets an LLM agent observe the live trading portfolio and open or close positions through guarded tools

Anthropic integrations for LangChain.js

Neo4j integrations for LangChain.js

Core LangChain.js abstractions and schemas

OpenAI integrations for LangChain.js