prioritize-assumptions
Pass
Audited by Gen Agent Trust Hub on Mar 4, 2026
Risk Level: SAFEPROMPT_INJECTIONNO_CODE
Full Analysis
- [PROMPT_INJECTION]: The skill processes user-provided files for context without implementing boundary markers or sanitization, creating a surface for indirect prompt injection.\n
- Ingestion points: User-provided files containing assumptions or research data (SKILL.md).\n
- Boundary markers: None defined; instructions and data are not delimited.\n
- Capability inventory: No executable scripts or tool-calling capabilities detected within the skill files.\n
- Sanitization: No input validation or instruction-ignoring wrappers are used.\n- [NO_CODE]: The skill is composed entirely of markdown instructions and metadata. It does not include any scripts, dependencies, or executable binaries.
Audit Metadata