product-data-cleanup
Pass
Audited by Gen Agent Trust Hub on Sep 27, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted external data which could potentially contain malicious instructions or shell commands.\n
- Ingestion points: The skill ingests data from multiple untrusted sources, including the
product-library.csvfile, user-provided CSV file paths, and markdown tables pasted into the session (as described in the 'Input' section of SKILL.md).\n - Boundary markers: The instructions do not define specific delimiters or security-focused boundary markers to isolate data from commands, nor do they explicitly instruct the agent to ignore instructions embedded within the data values.\n
- Capability inventory: The skill is configured with powerful tools such as
Bash,Write, andEdit. While intended for data manipulation, these tools increase the impact of a successful injection attack.\n - Sanitization: While the skill performs extensive semantic normalization (e.g., casing, unit conversion, and language translation), it lacks specific security sanitization to filter or escape content that could be interpreted as executable code or agent instructions.
Audit Metadata