canva-advanced-troubleshooting

Pass

Audited by Gen Agent Trust Hub on Sep 13, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and classify diagnostic evidence and incident records from external sources, which acts as an attack surface for indirect prompt injection.
  • Ingestion points: The skill's instructions and 'Support Escalation Template' describe processing 'redacted evidence,' 'incident records,' and 'response bodies' from API calls.
  • Boundary markers: The skill does not define explicit delimiters or instructions to the agent to disregard potential commands embedded within the diagnostic data being analyzed.
  • Capability inventory: The skill possesses network capabilities (using fetch and curl to communicate with api.canva.com) and file system access via the Read and Grep tools.
  • Sanitization: While the skill advises manual redaction of telemetry, it lacks automated mechanisms to sanitize natural language instructions that might be present in the evidence logs.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 13, 2026, 06:40 AM
Security Audit — agent-trust-hub — canva-advanced-troubleshooting