brand-guidelines
Pass
Audited by Gen Agent Trust Hub on Apr 11, 2026
Risk Level: SAFEPROMPT_INJECTION
Full Analysis
- [SAFE]: No malicious patterns, hidden code, or dangerous command executions were detected. The skill contains legitimate brand assets and instructional constants.\n- [PROMPT_INJECTION]: The skill provides a surface for indirect prompt injection by instructing the agent to process and style external, untrusted artifacts.\n
- Ingestion points: Triggered to process user-provided slides, documents, reports, and HTML artifacts as described in the metadata.\n
- Boundary markers: None present; instructions do not include delimiters to isolate brand guidelines from the artifact content being processed.\n
- Capability inventory: The skill does not define any direct capabilities such as subprocess execution, network access, or file-writing, which significantly limits the risk of an injection attack.\n
- Sanitization: No sanitization or input validation procedures for the external artifacts are mentioned in the skill instructions.
Audit Metadata