blog-post-writer
Pass
Audited by Gen Agent Trust Hub on Sep 15, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill exposes an attack surface by consuming untrusted text data. Ingestion points: User-provided text fields known as brain dumps enter the context in SKILL.md. Boundary markers: Absent; there are no clear boundary definitions or framing instructions to isolate user content from the primary prompt. Capability inventory: The skill is restricted to formatting and copyediting, with no execution hooks, subprocess commands, or risky API tools available. Sanitization: Absent; input validation or text escaping protocols are not implemented.
Audit Metadata