customer-feedback-analyzer
Pass
Audited by Gen Agent Trust Hub on Sep 25, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and synthesize customer feedback from multiple external sources, creating a potential surface for indirect prompt injection attacks.
- Ingestion points: Processes external data from sources like support tickets, NPS surveys, and feedback widgets as described in the Core Principle and Feedback Channels sections of SKILL.md.
- Boundary markers: The synthesis templates and guidelines lack explicit delimiters or instructions to treat external feedback as untrusted data, which could lead the agent to follow instructions embedded within the feedback.
- Capability inventory: The manifest.yaml indicates that the skill has the capability to modify files and create artifacts based on the results of the synthesis.
- Sanitization: No input validation, filtering, or sanitization logic is specified for the raw feedback strings before they are processed by the agent.
Audit Metadata