wasm-compatibility
Pass
Audited by Gen Agent Trust Hub on Sep 15, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes external marimo notebook files, which presents a surface where instructions embedded in a notebook could attempt to influence the agent's analysis.
- Ingestion points: The agent reads a target notebook file provided by the user (SKILL.md, Step 1).
- Boundary markers: There are no explicit instructions or delimiters defined to separate the notebook's code/content from the agent's operational instructions.
- Capability inventory: The skill uses the
marimo checkandmarimo exporttools via subprocess calls (SKILL.md, Steps 2 and 4). - Sanitization: No specific sanitization or filtering of the notebook content is defined before the agent analyzes the code for manual findings.
Audit Metadata