splitting-datasets
Pass
Audited by Gen Agent Trust Hub on Mar 12, 2026
Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill uses the
Bash(cmd:*)tool to execute Python code that it generates dynamically based on user prompts. This allows for arbitrary command execution within the agent's environment. The skill's primary function is to generate scripts for data partitioning at runtime.- [PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection through the datasets it processes. - Ingestion points: Data is ingested from user-provided files like CSV, JSON, and Parquet as described in
SKILL.mdandassets/dataset_schema.json. - Boundary markers: There are no explicit delimiters or instructions to the agent to ignore instructions embedded within the datasets.
- Capability inventory: The skill possesses
Bash,Read,Write, andEdittools as defined inSKILL.md. - Sanitization: No evidence of input validation or sanitization was found in the provided scripts or instructions to prevent malicious data from affecting the generated code logic.
Audit Metadata