alterlab-pra-consumer-insight
Pass
Audited by Gen Agent Trust Hub on Sep 22, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and process data from external, untrusted sources, creating a potential surface for indirect prompt injection attacks.
- Ingestion points: The workflow in
SKILL.mdexplicitly directs the agent to 'Search the web for current demographic data' and 'Read existing project files for context' when scoping research. - Boundary markers: The instructions do not define specific boundary markers or 'ignore' commands to help the agent distinguish between research data and instructional content.
- Capability inventory: The agent possesses capabilities to perform web searches, read local project files, and write new deliverables as markdown files (e.g.,
{project}-consumer-insight.md). - Sanitization: There are no specified procedures for sanitizing or validating the content retrieved from web searches or external documents before it is processed or used to generate new files.
Audit Metadata