Autonomous AI agents that perform tasks independently
ogabrielluiz
A turn-based, intent-level environment + scenario suite for evaluating LLM agents on FTL: Faster Than Light
genlayerlabs
An agent small enough to run anywhere. A minimal agentic runtime in C — ~380 lines, 54KB binary, ~2MB RAM. A skill file + an LLM + a shell loop, no framework.
...moreGT-Projects256
Open-source security firewall for MCP servers. Scan for OWASP MCP Top 10 vulnerabilities, enforce runtime policies on AI agent tool calls, and generate compliance audit logs.
...morezpb911km
基于生命周期钩子的轻量级 AI Agent 框架 — 插件即文件，全异步，零 SDK 依赖
calisweetleaf
Aeriadne is the first-of-its-kind marketplace for agentic operability. It delivers high-fidelity skills, plugins, subagent prompts, and server-plane activations to any coding agent (Codex, Claude Code, OpenCode, Grok Build). This contains one tool server(110+ tools),plugins, and more
...moreErtinox7711
⚡ The Claude Code power-rig in 1 prompt: settings.json, CLAUDE.md AGI-proactive, context hooks, memory, 12 plugins + an anti-leak pipeline of secrets to participate in the configuration without risk. 🇫🇷 FR-native.
...moremnemopay
Governed computer-use desktop agent for MnemoPay. Gives an AI its own containerized Linux desktop (see/click/type) where every action is policy-checked, risk-scored, budget-gated, approval-routed, and written to an Ed25519-signed Article 12 audit chain. T
...moredavcavalcante
Universal, drop-in, measurable LLM router with native streaming and tool calling. Routes prompts across OpenAI, Anthropic, Gemini and any OpenAI-compatible HTTP endpoint by cost, latency and observed quality. Includes a Vercel AI SDK adapter. Zero runtime
...morearmazarashi0107
Open-source Agent SDK inspired by @anthropic-ai/claude-agent-sdk. Runs the agent loop in-process — no local CLI required. Deploy anywhere: cloud, serverless, Docker, CI/CD.
...moreDetect stale facts in LLM agent memory stores
Shubham Divakar
Fit your whole repo into any small model's token window — context selection for free/small LLMs.
Islo
Islo (islo.dev) sandbox integration for Deep Agents
Halo host adapter for the OpenAI Agents SDK (Python): a call_model_input_filter that swaps large tool results for a shape map (out of context) plus a halo_fetch navigation tool. install_halo() wires it in one call.
...moreRun Claude Code agents in isolated, disposable sandboxes — one per task, with cheap rollback and a human gate.
the Ezra authors
Slim SDK for driving a deployed Ezra multi-agent platform over HTTP
A production-grade AI agent framework with 21 integrated tool classes and a four-role autonomous LLM pipeline — built on the NPMAI ECOSYSTEM.
...moreDeval Shah
Sift — deterministic website indexing for grep-first LLM agents
farhanic017
MCP vision server for AI agents, Claude, Cursor, OpenCode, VS Code, local LLMs, Ollama, LM Studio, OpenRouter, Gemini, image analysis, and video understanding.
...morekyle641320
An AgentSkill for Java enterprise backend work — guides coding agents to investigate the repository first, then plan, implement, and verify changes. Framework-agnostic, MIT.
...moreZWISERFIT
ZWISERFIT AI Store Manager — Hermes/OpenClaw agent system running gym operations 24/7. PoPB data pipeline + RWA tooling.
QT7-C23
14 skills that write every document your open-source project needs — README to GOVERNANCE, professionally and consistently.
...moreziroagent
ZiroAgent SDK — production-safe TypeScript agent runtime: durable execution, cost guardrails, replayable traces, MCP-native, sovereign-ready. https://ziroagent.com
...moreitsjwill
Persistent multi-agent Claude Code system with auto-restart, inter-agent task routing, structured memory, and Telegram control. Better than ClaudeClaw for autonomous AI agent teams. Free, open source.
...more