RepoRadar
Leave a Tip
Item detail

hud-evals/hud-python

hud-evals/hud-python is an MIT-licensed Python toolkit for building reinforcement-learning environments and evals for AI agents across coding, browser, computer-use, and robotics tasks, with one environment spec that runs as both held-out evals and online training loops.

Open source →Back to radar
Score8.4
Popularity6.5
Riskconditional
TierGold
Score breakdown
Usefulness8.0
Novelty8.0
Momentum7.0
Maturity6.7
Open-source/build8.4
Evidence7.2
Workflow potential9.9
Setup ease4.2

Popularity is tracked separately. Support, ads, sponsorships, and tips never affect these signals.

Why it matters

Useful for agent builders, eval authors, and RL researchers who want a single environment contract that scales from a smoke test to a multi-model training run.

Who should use it

agent teams running repeatable evals across Claude, GPT, and open-weight modelsRL researchers collecting trajectories from coding, browser, or computer-use taskseval authors who want one spec for both offline scoring and online trainingoperators triaging regressions before shipping agent releases

Who should skip it

Skip if the source link, docs, or setup requirements do not match your workflow.

Risk explanation

It executes agent code and may interact with live browsers or computer-use harnesses, so sandbox the runtime and review what data leaves the boundary during training.

Evidence links

Closest alternatives / related signals

  • No close related item is available in the current public catalog.
rlagent-evalgrpolorabrowser-agentstooling