content-engine
Pass
Audited by Gen Agent Trust Hub on Sep 17, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted source materials (such as articles, videos, or demos) to generate social media content and is granted file-writing capabilities. The instructions lack explicit boundary markers or sanitization steps for this external data, creating a surface for indirect prompt injection where malicious instructions embedded in the source could potentially influence agent actions.
- Ingestion points: Source content (articles, videos, demos) provided by the user as input for the 'Repurposing Flow' in
SKILL.md. - Boundary markers: Absent; the instructions do not specify the use of delimiters or provide warnings to the agent to ignore embedded instructions within the source material.
- Capability inventory: The skill is permitted to use the
Read,Write, andWebSearchtools as defined in the YAML frontmatter ofSKILL.md. - Sanitization: No input validation, escaping, or sanitization logic is described for the external content before it is processed.
Audit Metadata