Autonomous AI agents that perform tasks independently
Claude-Code-driving layer for Kagura harnesses — headless claude -p launcher, Anthropic auth, verdict, doctor primitives
Cairn contributors
A build system for LLM computation over a corpus of files
Agent OS Kernel — CPU-scheduling-inspired runtime for LLM agent workloads
A lightweight, scalable persistent memory system for LLM applications and long-term interactive agents
raybird.chen
Cross-session, traceable, self-hostable persistent memory for AI agents — SQLite-backed CLI + HTTP API + Node SDK.
ultirequiem
Installable skill packages for Requiems API — enables AI agents to interact with its endpoints natively.
yusufnb
Edit SleekCMS sites locally — models, content, templates, images — with live two-way sync and AI agent support.
eshaank08
Open source CLI that audits AI agents for security, performance, and model fit
cyb3rward0g
Agent-agnostic physical AI contracts for sessions, observations, capabilities, safety, and traces
rushikesh_g25
Cli coding Ai agent like Claude Code and Gemini Cli
Astra Framework - Core APIs and abstractions for building AI agents, teams, RAG pipelines, and workflows
vast limits GmbH
pySigma uAQL backend
Multi-model coding agent powered by NVIDIA Llama (3B, 70B) with CLI, web search, and tool-use loop
AgenticDome
Official Python SDK for AgentGuard Intelligence Engine and Action Firewall.
OpenAI Agents SDK tool for hard spending limits on AI agents via Valta
LangChain tool for hard spending limits on AI agents via Valta
CrewAI tool for hard spending limits on AI agents via Valta
iFlow Search SDK contributors
CrewAI tools for iFlow Search (心流搜索) — web search, image search, and web-page fetching.
Run ClawHub skills, sandbox, Playwright browser & multi-language runtimes as Hermes hands/feet. Structured JSON, zero extra LLM tokens.
...moreCarlos
Your agent proposes. Orca verifies. You decide. A drop-in verification layer for LLM/agent outputs.
Brain layer for Kagura harnesses — headless CLI coding-agent launcher (claude -p today), subscription-auth hygiene, verdict, doctor primitives
...moreHiClaw Contributors
HiClaw harness worker — delegates the agent loop to Claude Code, Gemini CLI, OpenCode, or Codex. Includes the harness-remote developer CLI.
...moreSuperagent safety integration for Hindsight - guard and redact memory operations
Three-tier memory with trust-aware context ledger for multi-agent LLM systems