video-script
Pass
Audited by Gen Agent Trust Hub on Oct 13, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes data from external files and user-provided links, creating a surface for potential indirect prompt injection attacks.
- Ingestion points: The skill reads brand configuration, profile data, compliance rules, and campaign briefs from the
~/.claude-marketing/directory (e.g.,profile.json,_manifest.json). It also accepts user-provided competitor video references. - Boundary markers: There are no explicit instructions or delimiters defined to prevent the agent from following commands that might be embedded within the ingested brand files or external references.
- Capability inventory: The skill is configured to read local files from the user's home directory (
~/.claude-marketing/). - Sanitization: The instructions do not specify any validation or sanitization steps for the content retrieved from brand profiles or competitor links before it is processed by the AI.
Audit Metadata