Search

Agentic Lab Skills

Self-curating memory for LLM agents: MeMo-style external memory kept honest by survival-based selection instead of reward models or judges.

LangChain & LlamaIndex tools for the Orbis API Marketplace — pay-per-call access to 8,000+ APIs via x402 USDC micropayments

A Tabby terminal plugin that adds an AI assistant side panel to terminal tabs, with chat, terminal context sharing, and tools for reviewing and running terminal commands. The command approval mode is configurable in settings, and the panel shows each comm

AI agent editor integration: diff UI, action registry, agent panel plugin

Portable repository-native foundation for agent memory, skills, handoffs, and knowledge workflows.

Top-level Graphlit tool factories for TypeScript streamAgent applications

OpenAI Agents SDK session memory backed by PerSQL — durable conversation history in an isolated SQLite database per agent.

Open-source verifiable agent accountability — cryptographic capability receipts and formal proofs for LLM agent activity

蓬莱 Penglai — 住在你飞书和微信里的中文 AI 管家 · Your personal AI butler in Feishu & WeChat (powered by GenericAgent, kernel untouched)

可嵌入的 AI 智能体可视化 Vue 插件 — 浮动按钮、流式对话、Markdown/图表渲染

AI Agent framework on LangChain/LangGraph — multi-backend filesystem, sub-agent scheduling, conversation summarization, security review, and skills system out of the box.

Skills extension for the OpenAI Agents SDK

Composable primitives for building LLM agents — atoms, protocols, molecules, rendering, organisms

Hono HTTP server for agentic-patterns agents — routes, SSE streaming, admin API

agentic — CLI for running agent projects. Crawls ./agents/, wires observability, serves chat + admin dashboard.

Agent-friendly local Odoo development runtime: worktree-isolated databases, ports, and Docker Compose stacks

Coding agent CLI with read, bash, edit, write tools and session management

Lightweight local AI coding agent — zero external dependencies, single process startup, with intelligent model routing (Claude > Qwen > Ollama).

OpenAI-Agents tools for the HonorNet agent presentation client — a protocol shim that mints proofs and presents three-part ADR-009 bundles. It decides nothing (hard rule 3).