#ai-agents

A streaming action runtime for AI agents, model serving, and multimodal APIs

Local-first shared memory for AI agents — zero-dependency, governed writes, citations, MCP. Source behind the ihow-memory npm package.

🧙 Grimoire is a skills package manager for AI agents and best practices — 1000+ skills across 26 domains (engineering, law, finance, design, health, and more), each sourced from a real institution or study. Install once, and your AI routes to the right skill automatically.

OpenCode Plugin for supervised agentic software development on a Kanban board

Audit agent for Claude Code that catches missed tool calls — separates spotting from doing. Your Claude answers, Spotter watches.

Cut ~90% of Claude Code's context usage while keeping nearly all the memory

A team of specialized AI agents covering the complete software development flow — from requirements discovery to deploy. One responsibility per agent, specs as source of truth, no shortcuts.

Agent-native RPC for the multi-agent era — write a Java interface, get gRPC + HTTP/MCP gateway and generated TS/Dart/Python/Rust clients, no protobuf. 智能体时代的契约先行 RPC：接口即 API，天然带 MCP 网关

AI venture studio: mines your support tickets and issues for opportunities, sizes markets honestly, writes PRDs with kill criteria, builds and reviews the product from one plain-language doc (English/中文), measures real cohorts, and forces honest kill decisions. Humans hold every gate; it never publishes or spends.

AISEF — AI Software Engineering Framework: one requirements file in, a gated and evidence-backed application out. Agents do the judgement; code enforces the guarantees.

Claude Code with Hebrew, Arabic and Persian rendered correctly. Most terminals have no bidirectional text engine, so right-to-left output comes out scrambled. This runs Anthropic's AI agent in a terminal that gets it right, with a folder picker instead of commands. Windows and Linux, free and open source.

Soma makes AI agents trustworthy by providing observably traceable governance, evidence-based rule selection W/ JIT, AST analysis tools, TTC Verifiers, and adaptive rule lifecycle management.

Go implementation of the Offering Discovery Protocol.

Yet another minimal and lightweight terminal multiplexer built for the agent era.

Agent factory workflow engine for agents

Infrastructure as Code for AI Agents. Declare MCP servers, skills, and hooks in gandalf.toml. Review before apply. Roll back safely.

WeChat/微信 local chat history MCP server for macOS agents. Read messages, contacts, media, favorites, transfers, red packets, Moments and full-text search from local WeChat 4.x data.

Require fresh, file-bound verification evidence before coding agents declare completion.

A math AI assistant agent

Drive your already-running local Chrome from the command line over the Chrome DevTools Protocol (Go + chromedp).

Generative-AI abstraction (LLM, agents, image/audio/video)

milk — a CLI/TUI that routes prompts between agents: a primary agent and a configurable escalation agent, with session-aware state management and real-time streaming. Escalation happens automatically (via routing rules) or manually, preserving context across the handoff between agents.

TUI for managing multiple AI coding agents in terminal tabs and panes.

Multi-client ACP session daemon: spawn coding agents, share live sessions across terminals, editors and browsers

Provider-agnostic AI SDK for Go — unified interface for building AI agents across LLM providers

Python, Javascript, and Rust libraries for the Spider Cloud API.

SLOP: Structured Language for Orchestrating Prompts - A domain-specific language for AI agent workflows

Model-agnostic design knowledge base + skill system for AI coding agents (Korean-market-fluent)

DNA by Humaan — project brain for AI coding. Doctor install, Lab, 1045-pack marketplace, feature factory, tagged commits/PRs per project.

Terminal trace debugger for AI agents.