Python functions powered by AI agents - with runtime post-conditions for reliable agentic workflows.
-
Updated
Sep 4, 2026 - Python
Python functions powered by AI agents - with runtime post-conditions for reliable agentic workflows.
Control robots and physical hardware with natural language through Strands Agents.
Strands-based agents and harnesses for agentic benchmarks.
Optimize LLM agent harnesses through tunable Formulas to dynamically enhance the agent, and improve those Formulas with optimizers based on collected agent rollout trajectories.
Simulated environments for robot agent evaluation and reinforcement learning.
Python bindings for WebRTC's audio processing pipeline, bringing noise suppression, echo cancellation, automatic gain control, and voice activity detection to the Python ecosystem via pybind11.
Give your agent eyes that understand physics and hands that generate video, audio, and robot actions.
To associate your repository with the strands-labs topic, visit your repo's landing page and select "manage topics."