Practical augments for human capability.
An augment is a tool that extends what a person can do. It does not replace the person or decide for them, and it will not demand their attention. The person stays in command.
Augments Labs is where we build them. Each project below has its own repository and its own docs.
Projects
Rust
Crucible Code
The harness where agents are forged.
Explore →Python
Augments ADK
A framework to orchestrate a complex system of agents that performs real-world actions.
Explore →Shell
SDLC Skills
Rigorous, phase-isolated SDLC skills that tether autonomous agents to real-world engineering standards.
Explore →What guides us
- Human authority
- A tool from this lab carries out what it was asked and leaves the decisions that matter to the person using it. It never acts on its own behalf.
- Understandable behavior
- You can see what a tool is doing and why at every step. If it goes somewhere you did not expect, you can stop it and take over.
- Meaningful choice
- Each tool works with the models, agents and services you already use. Nothing here ties you to one provider, and you can swap a part out when a better one comes along.
- Local ownership
- A tool from this lab runs where you run it, on your machine or in your own environment. Your work and your data stay with you.
- Measured usefulness
- A tool earns its place by what it does for the person using it. Finished work and fewer mistakes count for more than an impressive demo.