analyze-performance-diagnose
Pass
Audited by Gen Agent Trust Hub on Nov 8, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes data from potentially untrusted external sources, which creates a surface for indirect prompt injection.
- Ingestion points: The skill reads external market research files (
research/market/<D>-read.md) and visits destination URLs of advertisements. - Boundary markers: Absent. There are no explicit instructions to the agent to treat ingested data as data rather than instructions, nor are there delimiters specified.
- Capability inventory: The skill allows for file system reads and web browsing capabilities.
- Sanitization: Absent. There is no evidence of content sanitization or validation of the data retrieved from market research or web pages.
Audit Metadata