scribe
Pass
Audited by Gen Agent Trust Hub on Oct 13, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONEXTERNAL_DOWNLOADS
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill is designed to process user-provided text for stylistic refinement. * Ingestion points: User input drafts as defined in the
argument-hintofSKILL.md. * Boundary markers: Absent, though the skill provides instructions to filter content and stick to primary sources. * Capability inventory: Limited to text generation; the skill has no tool access and disables further model invocation. * Sanitization: No input sanitization is performed. - [EXTERNAL_DOWNLOADS]: References to external documentation and attribution are present. * Evidence: The skill links to Wikipedia and a public GitHub repository in
references/ai-tells.mdfor informational purposes. These links point to well-known services and do not involve executable code.
Audit Metadata