Dataset Curator
Pass
Audited by Gen Agent Trust Hub on Feb 16, 2026
Risk Level: LOWPROMPT_INJECTION
Full Analysis
- PROMPT_INJECTION (LOW): The skill exhibits an indirect prompt injection surface as its core functionality involves the ingestion and processing of untrusted external datasets.
- Ingestion points: Dataset files are accessed and profiled throughout Workflows 1, 2, and 3.
- Boundary markers: The skill does not provide instructions for the agent to use delimiters or ignore instructions embedded within the dataset samples.
- Capability inventory: The skill includes logic for data profiling, filtering, and modification using Python scripts.
- Sanitization: While the skill addresses data quality issues like duplicates and label noise, it lacks mechanisms to sanitize natural language content for adversarial instructions that could influence the agent's behavior during the curation process.
Audit Metadata