improving-prompts
Pass
Audited by Gen Agent Trust Hub on Oct 30, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONNO_CODE
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill ingests user-provided prompt text or files for optimization, creating an attack surface where instructions embedded in the processed data could influence the agent. Ingestion points: argument-hint and description in SKILL.md. Boundary markers: absent. Capability inventory: file-read and file-edit capabilities are referenced in SKILL.md. Sanitization: absent.
- [NO_CODE]: The skill consists entirely of natural language instructions and does not include any scripts, executables, or code-based logic.
Audit Metadata