canvas-course-qc
Pass
Audited by Gen Agent Trust Hub on Sep 25, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [SAFE]: Comprehensive analysis of the instructions and metadata shows no signs of malicious intent, command obfuscation, or unauthorized access to system resources. The skill uses appropriate Canvas-specific MCP tools to fulfill its administrative purpose.
- [INDIRECT_PROMPT_INJECTION]: The skill identifies a theoretical surface for indirect prompt injection by processing external course data and having access to write-capable tools. 1. Ingestion points: Assignment metadata, descriptions, and page content retrieved via MCP tools as specified in
SKILL.md. 2. Boundary markers: No explicit delimiters or instructions to disregard embedded content are provided in the skill logic. 3. Capability inventory: The skill can perform modifications usingupdate_moduleandbulk_update_pagesas defined inSKILL.md. 4. Sanitization: There is no evidence of specific sanitization or filtering logic for the ingested content. The inclusion of user confirmation for auto-fix actions serves as the primary security control.
Audit Metadata