set-note-description
Pass
Audited by Gen Agent Trust Hub on Sep 21, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted markdown content from user notes, which could theoretically contain instructions intended to influence the agent.
- Ingestion points: Reads note contents provided via arguments and linked component notes during hierarchical rollups.
- Boundary markers: Uses instructions to define the agent's role as a markdown analysis assistant and strictly limits output to the description property.
- Capability inventory: Performs file read and write operations on markdown frontmatter using standard agent tools.
- Sanitization: Explicitly instructs the agent to wrap generated summaries in double quotes and preserve all other formatting and syntax exactly as found.
Audit Metadata