Search

Temporary package placeholder for staged publishing

Temporary package placeholder for staged publishing

Regularized self-improvement of LLM agent harnesses: gains that hold up out of distribution, with verifiable change history

Tracely Python SDK: automatic + manual agent tracing, exported to Tracely over OTLP.

safe-agents — the trusted-autonomy base platform SDK (deterministic tool broker, agent pipeline, cloud arms, runner-contract harness).

Muse Wonder is a frontier reasoning model from Meta, designed for complex personal and professional agentic tasks. It accepts text, images, video, audio, and PDF documents and returns text, with...

An experimental platform separating information from enforcement in distributed LLM coding agents

CrewAI tools for EconDash macroeconomic data. Same free trial, API key and pay-per-call rules as the econdash client.

Agent primitives for Polychat: define an agent from a model, tools and an objective, run it through the bounded decision loop with approvals, and schedule it as a worker

Run coding agents (Claude Code, Codex, Vibe, pi) in rootless Podman sandboxes on Linux