composing-grid-canvases
Pass
Audited by Gen Agent Trust Hub on Sep 18, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and process potentially untrusted external data.
- Ingestion points: As described in SKILL.md, the agent reads user-provided feedback through
tasks-comments-listandtasks-comments-retrieve, and uses thepromptfield in a grid placement as the primary brief for its generation tasks. - Capability inventory: The skill utilizes powerful modification tools including
canvas-layout-patch,canvas-layout-publish, andcanvas-create, which allow it to alter the user interface and component logic (SKILL.md). - Boundary markers: The instructions do not include any guidance on using delimiters or safety markers to help the agent distinguish between its instructions and the data being processed.
- Sanitization: There are no specified sanitization or validation steps to ensure that content retrieved from external prompts or comments does not contain malicious instructions.
Audit Metadata