ds-experiment
Pass
Audited by Gen Agent Trust Hub on May 6, 2026
Risk Level: SAFEPROMPT_INJECTION
Full Analysis
- [PROMPT_INJECTION]: The skill possesses a vulnerability surface for indirect prompt injection.
- Ingestion points: SKILL.md instructs the agent to read 'shell logs and generated outputs from the actual run', 'metrics.json', and 'artifact_manifest.json' as part of its experimental workflow.
- Boundary markers: There are no explicit instructions or delimiters used when reading these files to ensure the agent ignores any malicious instructions that might be embedded within the logs or metrics.
- Capability inventory: The skill has extensive capabilities, including executing arbitrary shell commands via 'bash_exec(...)', managing repositories via 'artifact.git(...)', and recording results through 'artifact.record_main_experiment(...)'.
- Sanitization: No sanitization or validation of the ingested content is performed before the agent uses it to generate summaries and determine the next course of action.
Audit Metadata