Autonomous AI agents that perform tasks independently
nczz
Pi-native execution contracts for AI agents — evidence-gated completion, ordered plans, and compaction-safe resume
xfajarr
Thiny AI — a beautiful terminal agent: interactive chat, tools, Walrus memory, and Sui execution.
GitHub Actions
On-demand local gate for your messages. Reveals WHO contacted you; only reveals WHAT for approved senders. iMessage today, pluggable channels next.
...moreGitHub Actions
Captatum — adaptive MCP web-fetch tool for AI agents: fetch any URL, render JS when needed, return clean structured content + provenance.
...moreappcypher
Lightweight VM sandboxes for Node.js and TypeScript applications that need hardware-level isolation for AI agents, tools, tests, and untrusted code.
...moreChandler Song <[email protected]>
Typer + Rich CLI for recruit-loop-agent — run recruiting loops from the terminal
Rawson Consulting B.V.
JWT-gated LLM gateway: authenticate (bcrypt/JWT), then run a LangChain-on-Vertex Gemini completion. Unauthenticated calls are rejected.
...moreLangChain and LangGraph integration for GaaS (Governance as a Service)
ManulEngine Contributors
Deterministic DSL-first browser automation runtime driving Chrome over the DevTools Protocol (CDP) — for humans and LLM agents
...moremmprotest
Some tools help. Some tools assist. Villani Code intervenes.
JosephHu04
🏨 五星级酒店客房服务 AI Agent — ReAct + LangGraph + Ollama 本地 LLM | v1 基础框架 → v2 云端 DeepSeek → v3 本地化零成本 | 行业无关的通用 Agent 骨架
Kaushal-Panchal
⏱️ Time-boxed, self-improving research for Hermes, say "research X for 4 hours," walk away, get a cited report.
OpenAI Agents SDK integration for GaaS (Governance as a Service)
BUJJI - The first truly local-first AI Agent SDK. Run powerful agents on your laptop with Ollama, AirLLM, and 13+ built-in tools.
...moreGrid-based LLM agent simulation engine — Session API, multi-area worlds, tactical movement, compound turns, pluggable memory
...moresocialsyncs
SocialSyncs CLI - Command line interface for the SocialSyncs social media scheduling API
robotrock
Human-in-the-loop approval workflows for AI agents
GitHub Actions
Pi extension: Claude Code-style subagents. Delegate tasks to isolated, per-agent provider/model-configurable specialists (general-purpose, explore, plan) defined as Markdown files, driven entirely through tools (no TUI required).
...morePython SDK for MakerChecker: govern AI-agent tool calls with deny-by-default grants, segregation of duties, and a hash-chained audit trail.
...moreStateless, subprocess-isolated code execution for LLM agents
antgroup
Mobile-native agent SDK for building on-device AI agents with sandboxed execution, extensible tools, and x-app / x-agent / x-channel connectivity.
...morebzcsk2
Free-first, model-adaptive coding agent runtime that makes local small LLMs work reliably with harness, checkpointing, verification, cache-aware context, and free-model fallback.
...morekevics1
三层地理空间认知系统：感知(L1)→理解(L2)→推理(L3)。18 MCP工具 + 3 Skills + 领域插件，为LLM Agent提供空间感知与因果推理能力。