termigen/termigen-environments
Coding
TermiGen-Environments: verified Docker environments with executable terminal-agent tasks across 11 categories, generated by an end-to-end multi-agent synthesis pipeline.
Run this task
CLI:
inspect eval inspect_harbor/termigen_environments --model openai/gpt-5Python:
from inspect_ai import eval
from inspect_harbor import termigen_environments
eval(termigen_environments(), model="openai/gpt-5")Dataset information
| Harbor registry | termigen/termigen-environments |
| Inspect task | termigen_environments |
| Latest digest | sha256:492c3b4c051b304b3887ca4a94a3081094c177b1227f0a609123da236359d5f0 |
| Samples | 1000 |
| Paper | arxiv |
| Source | https://github.com/ucsb-mlsec/terminal-bench-env |
See Task Parameters for the parameter set shared across all Harbor tasks.