Autonomous AI agents that perform tasks independently
hec-ovi
Agentic control layer for the Censurado self-hosted AI news portal: a stdlib Python CLI an LLM agent tool-calls to write and publish AI-persona articles through a gated editorial walk, plus the prompt/skill recipe and a 24/7 serve loop (Telegram bridge + agent fallback chain: Gemini, Codex, Claude, local model). Holds no data, runs no model.
...morereic
四工作流個人知識管理框架（wiki/project/trend-wiki/personal-km）——讓 Claude Code/Codex/Antigravity 當你的知識編譯器與簿記員；純 markdown 空骨架＋一鍵安裝包
Raghav Mishra (PowerAI Labs)
Record real LLM calls once, then replay them in your tests forever — offline, no API key, no cost, and the same result every run.
...moreGitHub Actions
FAi Brilliance — FieldAssist skill standards: browse on the web, install from any terminal or AI agent.
GitHub Actions
Legion — the model-agnostic orchestration engine under the hood of an AI agent: models, harnesses, loops, self-improving agents. The stack, not another wrapper. Scoped multi-model delegation (Codex, Cursor, Claude), telemetry, doctor/self-learn/heal loops
...morepatatatatat
Local supervision daemon for the Founder Control System — installs Claude Code hooks into a project and relays structured build events to a self-hosted FCS dashboard.
...moremelnixdev
OpenCode plugin for AI coding agents that caches safe repo context, code maps, and compact tool outputs.
soumyasurana
Contextra is a production-grade AI context engineering platform built in Rust for document ingestion, retrieval, embeddings, memory, prompt orchestration, and multi-LLM application development.
...morefablerlabs
An open-source operating system for long-lived, autonomous, revenue-generating agent businesses — memory, money, and governance for any LLM brain. Built and run by an autonomous AI agent.
...moreCoding-Dev-Tools
Framework-agnostic orchestration layer for AI agents: circuit breakers, multi-LLM council, workflow FSM, sweeps. Companion to Engraphis.
...moreAPL and J in notebooks: Iverson-language magics and kernels for Jupyter, IPython, and LLM agents
dan1510123
A 2D pixel-art desktop game where you walk around a virtual office and interact with AI agents that run real GitHub Copilot CLI sessions — plan, debug, orchestrate fleets, and drive agents from Microsoft Teams.
...moreLayered Invariant-Preserving Agent System — side-effect-aware, budget-gated, replayable LLM agent runtime
CampAIgn — map-based LLM RPG simulation engine (Session API, multi-area worlds, compound turns)
SiyaoZheng
GEZHI (格致): a scholar-centered agent framework built on goal-cli and ai4ss-skills for completed empirical social-science research.
...moreKuaishouGameMind
TRACE Bench: Task-driven Roleplay Agentic Checklist Evaluation
OpenTelemetry-native observability for multi-agent LLM systems
Prakhar Tripathi
Predict LLM-agent-run failures in real time — and intervene before the tokens are spent
Almanac
A provider-neutral harness SDK for agent systems on Claude and Codex.
ethedla
Keep your secrets on a leash — AI agent skill that catches exposed API keys, tokens, and credentials before they hit your codebase
...moreAugment Code Inc.
TypeScript client for the Auggie v2 SDK / RPC protocol
Vinctor contributors
Deterministic authorization core for mediated AI-agent actions.
Raj Zalavadia
Linter for AI instruction files — skills, prompts, and agent specs
Yihe Huang
Hardened single-node registry service for Agent Auth identities