weather-transformer
Pass
Audited by Gen Agent Trust Hub on Mar 1, 2026
Risk Level: SAFEPROMPT_INJECTION
Full Analysis
- [PROMPT_INJECTION]: The skill is vulnerable to indirect prompt injection because it explicitly instructs the agent to read and follow 'transformation instructions' from an external file.
- Ingestion points: Transformation rules are read from
weather-orchestration/input.mdusing the Read tool. - Boundary markers: No boundary markers, delimiters, or safety warnings are present to distinguish data from instructions within the input file.
- Capability inventory: The agent has access to Read and Write tools, allowing it to process the input and write results to
weather-orchestration/output.md. - Sanitization: There is no evidence of validation or sanitization for the content retrieved from the input file, meaning a malicious user or process could inject instructions to divert the agent's behavior.
Audit Metadata