optimize-prompt
Pass
Audited by Gen Agent Trust Hub on Mar 9, 2026
Risk Level: SAFEPROMPT_INJECTION
Full Analysis
- [PROMPT_INJECTION]: The skill incorporates untrusted user data into its logic, creating a surface for indirect prompt injection.
- Ingestion points: User descriptions are ingested via the
$ARGUMENTSvariable inSKILL.md. - Boundary markers: The skill template lacks delimiters or instructions that would tell the agent to ignore instructions embedded within the user's input.
- Capability inventory: The skill has access to the
Readtool (used to fetch reference documents) and theAskUserQuestiontool. - Sanitization: No input validation or sanitization is performed to prevent the agent from potentially obeying instructions hidden within the user-provided goals.
Audit Metadata