brain-dump
Pass
Audited by Gen Agent Trust Hub on Nov 12, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes raw, unstructured user input (thoughts, transcripts, notes) to categorize and summarize them into structured Markdown files. This creates a standard surface where malicious instructions embedded in the input could attempt to influence the agent. However, since this is the primary purpose of the skill, the severity is minimal.
- Ingestion points: Raw user input is ingested as described in the
SKILL.mdRoles & Responsibilities section. - Boundary markers: The skill does not explicitly use delimiters or instruction-ignore warnings when handling the input during the categorization phase.
- Capability inventory: The skill uses the
write_to_filetool to save notes to a local path specified by the user inSKILL.md. - Sanitization: No specific sanitization or filtering logic is present for the input text before it is formatted into the final note.
Audit Metadata