create-briefing
Pass
Audited by Gen Agent Trust Hub on Sep 29, 2026
Risk Level: SAFENO_CODEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes session data and research findings to generate new documents, creating an attack surface where untrusted input could be persisted in output files.
- Ingestion points: Content extracted from work sessions and research (SKILL.md, Step 1).
- Boundary markers: Employs a dual-audience structure (Human/AI sections) in SKILL.md acting as a logical delimiter, but lacks explicit instructions to ignore embedded instructions.
- Capability inventory: Writes output to the local file system (SKILL.md, Step 5).
- Sanitization: Input content is formatted into templates in SKILL.md without automated validation or escaping of the text content.
- [SAFE]: File operations are restricted to the local docs/briefings/ directory using standard naming conventions.
- [SAFE]: External references are limited to reputable documentation resources such as Atlassian, Asana, and Amazon templates.
Audit Metadata