business-growth-skills
Pass
Audited by Gen Agent Trust Hub on Sep 3, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes user-supplied JSON data (such as RFP requirements or customer health metrics) to generate analysis reports. While this presents a surface for indirect prompt injection, the risk is mitigated as the scripts perform numeric and algorithmic scoring rather than interpreting data as natural language instructions.
- Ingestion points: The skill reads JSON data via scripts like
rfp_response_analyzer.pyandhealth_score_calculator.py. - Boundary markers: No explicit delimiters are used in the instructions to isolate user-provided data.
- Capability inventory: The skill has access to terminal execution and local file systems for analysis.
- Sanitization: The logic is primarily mathematical, reducing the likelihood of malicious string interpretation.
Audit Metadata