cross-poster-blurb
Pass
Audited by Gen Agent Trust Hub on Sep 3, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill provides structural and stylistic guidelines for text generation. No malicious patterns, command executions, or network requests were identified.\n- [INDIRECT_PROMPT_INJECTION]: The skill involves processing external context (spine, voice-profile, audience-notes) to generate a blurb. While this defines an ingestion surface for potentially untrusted data, the skill lacks any capabilities that could be exploited via injection, such as shell access, file-system writing, or network exfiltration. The risk remains at a baseline level for text-generation tasks.
Audit Metadata