figure-description
Pass
Audited by Gen Agent Trust Hub on Nov 10, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONPROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes potentially untrusted external data from patent documents and figure files, creating an attack surface for indirect prompt injection.
- Ingestion points: Reads
patent/INVENTION_DISCLOSURE.md,patent/CLAIMS.md, and various figure files (PNG, JPG, SVG, PDF) discovered in the project directory. - Boundary markers: Absent. The skill does not define specific delimiters or instructions to the agent to disregard commands embedded within the processed patent files.
- Capability inventory: The skill is granted extensive capabilities, including
Bash(*),Write,WebSearch, andWebFetch, which could be abused if an injection is successful. - Sanitization: Absent. There is no evidence of content validation or sanitization for the ingested patent data.
- [PROMPT_INJECTION]: The user-controlled variable
$ARGUMENTSis directly interpolated into the instructions without sanitization or delimiters. A malicious user could provide input designed to override the skill's intended behavior or instructions.
Audit Metadata