Dabstep (Stalkermustang)

Data Agent Benchmark for Multi-step Reasoning benchmark

Type: RL Env
Tags: Data Analysis Openai Compatible Coding
Runtime: tool-use
License: unknown
Size: v0.1.0
Published: Sep 2025
Updated: Oct 2025
Canonical: app.primeintellect.ai/dashboard/environments/stalkermustang/dabstep

Cite

Notes

Only stored in your browser.

Attribution

README: api.primeintellect.ai/api/v1/environmentshub/stalkermustang/dabstep/@0.1.0/inspect

Evals this tool implements

Same problem set, this tool's harness. Run it to score a model on the test.

Hugging Face

450+ multi-step data-analysis tasks combining structured records with unstructured docs, requiring an agent to plan, query, and reason in stages.