Search

Provider-agnostic LLM adapter with streaming and multi-turn support

Minimal stateful agent runtime for agentic-kit

React bindings for agentic-kit (useChat hook)

Adversarial multi-agent development harness: PRD -> plan -> phased implementation, every artifact reviewed adversarially

Modern multi-agent CLI orchestrating multiple AI models via OpenRouter

A lightweight Python SDK for building fault-tolerant LLM agent workflows. It enables agent systems to recover from crashes by replaying execution from checkpoints, without re-invoking deterministic operations like LLM calls or completed tool invocations.

The universal entrypoint to HuggingFace diffusers for Strands agents — 100% pipeline & modality coverage, zero hardcoding. Special focus on Physical-AI world-foundation models (Cosmos) with robot action outputs.

HumanEval for tool use — a standardized benchmark for evaluating LLM tool-use capabilities

macOS desktop automation sandbox for AI agents

Personal AI agent toolkit installer for Claude Code, Codex CLI, OpenCode, Gemini CLI, Antigravity and graph-aware workflows.

Create a new BackendKit agent project

Drop-in observability SDK for AI agents — trace workflows, tools, retrieval and LLM calls, built on OpenTelemetry.

A ContextOps toolkit for AI coding agents.

openHermit: Loop Engineering workbench for local AI agent runtimes.

Subagent dispatch with live TUI streaming and cost tracking

Config-driven orchestration agent — decomposes tasks and coordinates specialist agents via shared vault and configurable LLM providers.

Merge your codebase into one text block for AI agents. Stop copying files manually.

Code diff assistant for VS Code.

Code diff assistant for VS Code.

Read-only CLI for a research data API (wiki, graph, memos, raw, coverage, eval). Self-documenting; an agent can drive it without a skill.