crud-tester
Pass
Audited by Gen Agent Trust Hub on Jul 30, 2026
Risk Level: SAFEPROMPT_INJECTION
Full Analysis
- [PROMPT_INJECTION]: The skill possesses an attack surface for indirect prompt injection by processing external data to drive agent actions.- Ingestion points: Workflow step 1 indicates the skill reads scenario lists in CSV or prose format to define its testing goals.- Boundary markers: The instructions lack specific delimiters or warnings to ignore potentially malicious instructions embedded within the ingested scenario data.- Capability inventory: The skill utilizes browser automation (Playwright) and file system operations, including reading target profiles and writing evidence logs and reports.- Sanitization: No sanitization or validation logic is defined for the external scenario content before it is processed by the agent.
Audit Metadata