speed
Pass
Audited by Gen Agent Trust Hub on Oct 27, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted text from user arguments or previous conversation history, which presents a potential injection surface.\n
- Ingestion points: Text is retrieved from $ARGUMENTS or previous conversation history in SKILL.md.\n
- Boundary markers: The instructions mandate JSON serialization and the escaping of HTML-sensitive characters (e.g., < as \u003c) to prevent the data from being interpreted as code.\n
- Capability inventory: The skill uses Write, Bash, and Read tools to generate and open a local preview file.\n
- Sanitization: There are explicit instructions to avoid passing source text to shell arguments and to ensure the content is treated as data, not instructions.
Audit Metadata