Run Your First Task¶
This walkthrough uses public task instances and produce-only mode—the same path an external benchmark runner uses before submitting results for official scoring.
Prerequisites¶
- Install ASI-Bench
- prepare an agent command that reads a task workspace and writes the requested outputs
- use Linux for the isolated
linux_nscustom-command path, or usenoneonly for trusted local testing
Step 1: Download Task Instances¶
The downloaded directory contains public prompts and input data, not private scoring material.
Step 2: Run Your Agent¶
asibench run --no-score \
--instances-dir hf_instances/ \
--agent-cmd 'python my_agent.py --workspace {workspace}' \
--sandbox linux_ns \
--output-dir out/
Replace the example --agent-cmd with your own non-interactive command. The {workspace} placeholder is replaced with the prepared task directory for each run.
| Option | Purpose |
|---|---|
--no-score |
Produce artifacts without access to private reference answers |
--instances-dir |
Read downloaded prompts and input data from this directory |
--agent-cmd |
Launch your agent against each prepared workspace |
--sandbox linux_ns |
Isolate a custom file-exchange agent with Linux namespaces |
--output-dir |
Store outputs and run provenance for later submission |
Step 3: Check the Output Directory¶
Confirm that out/ contains run metadata and the outputs requested by the task instances. Missing declared files will be reported when the bundle is uploaded.
Step 4: Submit the Run¶
Open the returned link and confirm the draft after reviewing its integrity and completeness summary.
Next Steps¶
- Run with Your Agent for built-in adapters and custom commands
- Understand Results for output files and provenance
- Submit Benchmark Results for the complete upload and review flow