Run from a terminal
The truagent command connects a prompt to an OpenAI-compatible model.
A compact Python agent toolkit
From a prompt to a tool call.
A Python agent library and command-line runner with no third-party runtime dependencies.
TruAgentPROJECT OVERVIEW
01Choose a provider
02Set a step limit
03Run a task
Illustrated workflow · not a live application session
01 / Inside the project
Built around the work.
Start with the implemented workflow, then follow the source to understand how it works.
Build with Python
Tool registration, memory, parsing and middleware are available as modules.
Keep the core small
The runtime uses the Python standard library; development tools are optional extras.
02 / Explore the workflow
Three steps to understand it.
Run from a terminal
The truagent command connects a prompt to an OpenAI-compatible model.
Full setup and implementation ↗03 / Run it yourself
Start from the source.
Python 3.9 or later. Configure OPENAI_API_KEY for a hosted endpoint or choose a local --base-url and --model.
Clone the repository and run these commands from its root. Review the README before connecting external services.
Open repository ↗SOURCE CHECKOUT
git clone https://github.com/honeyamn10-source/truagent.git
cd truagent
python -m pip install .
truagent version
truagent run "Explain what a Python iterator does" --max-steps 404 / Know the scope
Clear expectations.
Install from this repository. A package version in the source is not proof of a published PyPI release. Provider requests can incur costs.
Check the latest results in GitHub Actions. A passing automated check covers its tested paths, not every real-world integration.