Search

Agent Harness for Large Language Model Agents: A Survey. Survey on LLM agentharnessengineering with a taxonomy. 110+papers, 23 systems analyzed.

Empirical benchmark comparing agent architectures (single-agent, multi-agent, adaptive) on ProgramDev-v0 and CyberGym tasks. Key finding: adaptive architecture > single-agent > fixed-pipeline multi-agent.

Multi-agent orchestration runtime with task board, flow engine, budget control, MCP integration and real-time dashboard. Self-hosted on Linux/macOS.

A composite ai framework for agentic AI

Feed LLMs financial data and interactive charts in one call — an MCP server for A-share quotes, financials, and macro indicators, 42 tools with results-as-UI.

Drop in deal documents → get an onboarding plan, draft invoice, and stakeholder summary. Multi-agent LangGraph pipeline with RAG, human approval, and self-correcting retries.

Aglet tool technique that wraps another agent.yaml as a callable Tool — the multi-agent primitive.

Aglet safety technique: constrain agents with a declarative list of principles, enforced by an LLM judge.

LangChain integration for Chio protocol tools with conformance tests

CrewAI integration for the Chio protocol -- capability-scoped tools, per-role scoping, delegation attenuation

AutoGen integration for the Chio protocol -- capability-scoped function registration, GroupChat role scoping, nested chat attenuation

Security auditing CLI for AI agent skills — OWASP Agentic Skills Top 10 (AST10)

Give every AI agent an inbox - encrypted agent-to-agent messaging CLI with permanent addresses, durable threads, JSON automation, and human approvals.

OpenClaw plugin for long-term memory via OpenViking. Hook-only design, works alongside LCM without conflict.

OpenCode plugin for AKM - search, show, and manage extension assets via the akm CLI, with agentic hooks that auto-load relevant stash assets, record feedback, and harvest session memories so the stash improves every session.

Ultra-fast, 50KB browser vector database powered by Rust + WebAssembly. 96% RAM reduction via 1-bit quantization, HNSW indexing, OPFS persistence.

JavaScript/TypeScript SDK for MyMedi-AI healthcare APIs — medical coding, drug intelligence, reimbursement, clinical trials, physician payments, disease surveillance, prior auth, NER, claims, compliance. Zero dependencies.

CLI to manage and deploy AI agent skills

Product design skills for AI agents

Self-hosted Claude AI agent framework — gateway, memory, tools, Telegram, cron, heartbeat. Full system access via tool_use.