Search

Crucible. The touchstone — benchmark and verification tool for multi-agent dispatch.

Local-first software factory: deterministic multi-agent orchestration for small local models. Early development.

Agent + CLI that turns Claude into a private Spanish tutor: track progress, get an overview, get advice, read stories, and learn & practice Spanish (written and spoken) online from your phone.

Agent + CLI that turns Claude into a private French tutor: track progress, get an overview, get advice, read stories, and learn & practice French (written and spoken) online from your phone.

A minimal generic agent-loop framework on top of LangGraph: build_agent_graph assembles a standard ReAct loop (ai / tool / history / compaction / finish) with dependency-injected llm, checkpointer, tools, and runtime_context.

Testing-shaped eval framework for Agentick. defineEval(...) returns a callable with runtime parameterization; .matrix(axes) runs cartesian parameter sweeps — e.g. the same document/expected pairs across multiple models — and reports per-cell results.

CncertAgent local WSS log cache, offline analysis, submission, and API adapter toolkit.

Contract library for Sage platform agents

Scans your machine for dead side projects, autopsies why each one died from its git history, and resurrects the one with a pulse

A multi-agent ride-pooling & dispatching simulation environment for transportation gig-market research.

Deterministic, local-first observability for RAG and agentic pipelines (umbrella package + analyst CLI)

Deterministic workflow topology enforcement for LLM-powered systems.

Agent-driven hyperparameter optimization: LLM agents (claude / codex / opencode) as samplers and pruners.

Repo-agnostic Agent Skill evaluation harness with paired variants, holdout splits, repeated-run stats, script assertions, judge command backends, Anthropic-compatible exports, Jetty adapter support, and static review output.

Homekit gives any AI agent direct, physical control over Apple Home. Flip lights. Lock doors. Read sensors. Through a single open interface.

Give your AI assistant a phone — OpenClaw plugin for real phone calls via Twilio + OpenAI Realtime, with in-call tools, transcripts, and call screening

AI-powered MCP proxy for Playwright and Figma. Optimize Claude Code and AI agents with session recovery, context reduction and intelligent tool routing.

Agent OS: keep specialist agents in a hub, spin up a temporary orchestrator per task. Local-first, works with any model.

Open-source multi-tenant agent runtime for production. Tenant isolation, per-user cost controls, dual-sandbox security, deployment from Git.

Cloud-first AI agent via NVIDIA NIM Free API — rate limiting, key rotation, smart fallback chain. Local mode for offline use.