canvas-accessibility-auditor
Pass
Audited by Gen Agent Trust Hub on Sep 15, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill ingests and processes untrusted HTML content from Canvas pages and assignments, which could theoretically contain malicious instructions designed to influence the agent's behavior during the audit or remediation process.
- Ingestion points: Course content is retrieved from the Canvas API via
scan_course_content_accessibilityandget_page_contentas defined inSKILL.md. - Boundary markers: The instructions do not define specific delimiters or instructions for the agent to ignore natural language commands found within the HTML content it retrieves.
- Capability inventory: The skill has write capabilities to the LMS via the
edit_page_contenttool, which is used to apply accessibility fixes. - Sanitization: There is no evidence of content sanitization or instruction filtering applied to the retrieved HTML before the agent analyzes it or suggests remediation actions.
Audit Metadata