gbp-advisor
Pass
Audited by Gen Agent Trust Hub on Sep 26, 2026
Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: User-supplied variables such as business name, city, and report name are interpolated directly into shell command strings (e.g.,
python scripts/dfs_client.py business --keyword "<business name> <city>"). This creates a risk of command injection if the input is not sanitized before execution. - [INDIRECT_PROMPT_INJECTION]: The skill ingests and analyzes untrusted data from external sources, specifically Google Business Profile reviews and Q&A content, which are user-controllable.
- Ingestion points: External business data is retrieved via
scripts/dfs_client.pyas described inSKILL.md. - Boundary markers: The instructions lack delimiters or explicit warnings to the agent to ignore instructions embedded within the retrieved business data.
- Capability inventory: The skill has the ability to execute python scripts (
dfs_client.py,report_publish.py) and potentially write or read files during the report generation process. - Sanitization: No sanitization or validation logic is specified for the external data before it is processed by the agent.
Audit Metadata