proof-design
Pass
Audited by Gen Agent Trust Hub on Jul 1, 2026
Risk Level: SAFEPROMPT_INJECTIONCOMMAND_EXECUTIONEXTERNAL_DOWNLOADS
Full Analysis
- [EXTERNAL_DOWNLOADS]: The skill utilizes
WebFetchandWebSearchto access external web surfaces for design auditing. This is a primary function and is used to retrieve the visual elements being audited. - [COMMAND_EXECUTION]: The skill uses
Bashto generate reports and format CLI output. The instructions specify a 40-line CLI limit and the use of box-drawing characters for the report skeleton. - [PROMPT_INJECTION]: The skill possesses an indirect prompt injection surface as it processes untrusted data from external sources while having access to sensitive tools.
- Ingestion points: External URLs and visual surfaces fetched via
WebFetchas described in Phases 1 and 2. - Boundary markers: The instructions do not define specific delimiters or warnings to ignore embedded instructions within the fetched content.
- Capability inventory: The agent is equipped with
Bash,WebFetch,WebSearch, and file-reading tools (Read,Glob,Grep). - Sanitization: There is no mention of sanitizing or validating the content retrieved from external sources before processing.
Audit Metadata