pricing-pocket-waterfall
Pass
Audited by Gen Agent Trust Hub on Sep 8, 2026
Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill uses the
Bashtool to execute a local Python script (scripts/pocket_price_waterfall.py) intended for data processing and analysis. - [INDIRECT_PROMPT_INJECTION]: The skill contains an attack surface for indirect prompt injection due to the ingestion and processing of untrusted external data.
- Ingestion points: The skill reads external data from
customers.csv, which includes user-controlled fields such as customer IDs, discount reasons, and JSON-formatted discount steps. - Boundary markers: There are no explicit delimiters or specific instructions for the agent to disregard potential commands or malicious formatting within the ingested CSV data.
- Capability inventory: The skill is configured with
Read,Write, andBashcapabilities, which could be abused if the agent follows instructions embedded in the data. - Sanitization: The skill instructions do not include steps for validating, escaping, or sanitizing the input data before it is incorporated into the cohort-level waterfall reports.
Audit Metadata