Multi-agent LLM system for autonomous code generation, testing, and debugging with Docker sandboxing
AgentForge - Self-Evolving Agent Orchestration for Claude Code | Forge your workflow, evolve your agents
Java-first open-source AI Agent framework & workbench. Core framework: general-purpose ReAct agents and harness agents (sandbox, HITL approval, resumable execution). AgentForge Service: best-practice standard agent infrastructure to build, debug, publish and operate production-ready agents — streaming, ChatMemory, MCP/HTTP tools, session replay.
Linux of AI is a free, open-source ecosystem for building portable, governed, measurable, and replaceable AI infrastructure. It connects tools for agent orchestration, private deployment, model switching, policy, ontology, audit logs, and AI cost measurement to help reduce vendor lock-in.
MCP server for AgentForge — turn rough requests into tool-tuned, quality-checked prompts for AI coding agents.
CLI and TUI for managing, categorizing, and syncing reusable skills across AI coding agents and project workspaces.
AgentForge — Self-improving autonomous multi-agent system with Pure Rust Flywheel, task orchestration, and agent-driven development
LIBA AgentForge Day 1 Checkpoint 2: composable agent service and Streamlit interface
LIBA AgentForge Day 1 Checkpoint 3: reviewed multimodal evidence, memory and channels
AgentForge (Phase 2 of 3) - Multimodal Evidence Agent; document ingestion with local vision extraction, hybrid RAG + rerank, supervisor/worker orchestration, and citation contracts on an OpenEMR base. Continues agentforge-1-clinical-copilot at v1.0.
Enterprise knowledge assistants with NLU, RAG, omnichannel delivery
LangGraph, CrewAI, AutoGen with portable multi-cloud deployment
A Next-Generation, In-Browser Autonomous AI Coding Environment powered by WebContainers, Next.js, and Puter.js.
Canonical test rig for AgentForge's plugin system. Deterministic, zero-dep — the pre-release smoke gate.
AgentForge (Phase 1 of 3) - Clinical Co-Pilot Foundation; a local, verification-backed AI agent embedded in OpenEMR.
RAG pipelines, NL-to-SQL, hybrid retrieval with re-ranking
Multi-step ReAct agents with tool-calling and self-correction
React SPA front-end for AgentForge. Streams the agent's think/act/observe loop, with every mode, tool call, and connector in one chat UI.
AgentForge (Phase 3 of 3) - Adversarial Security & Red-Team Platform; a multi-agent platform that attacks the Phase 2 Clinical Co-Pilot.
To associate your repository with the agentforge topic, visit your repo's landing page and select "manage topics."