13-campaign-brief
Pass
Audited by Gen Agent Trust Hub on Oct 11, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONNO_CODE
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill ingests data from external sources (user input and outputs from other skills) which could contain malicious instructions designed to influence the agent's behavior.
- Ingestion points: The skill collects several data points in the 'Required Input' section (e.g., Campaign name, Target persona, Key message) and references outputs from
09-campaign-strategy,03-audience-personas, and04-brand-voice-intake. - Boundary markers: There are no instructions provided to wrap these inputs in delimiters (e.g., XML tags, triple backticks) or to include system instructions that explicitly warn the agent to ignore any embedded commands within that data.
- Capability inventory: The skill's functionality is limited to structured text generation in markdown format. It does not perform shell command execution, file system writes (outside of the workspace output), or network operations.
- Sanitization: No sanitization, escaping, or validation logic is defined to check the content of the provided inputs before they are interpolated into the document structure.
Audit Metadata