boring-email-sequence
Pass
Audited by Gen Agent Trust Hub on Mar 9, 2026
Risk Level: SAFEPROMPT_INJECTION
Full Analysis
- [PROMPT_INJECTION]: The skill ingests content from external files to inform its email generation, representing an indirect prompt injection surface.
- [PROMPT_INJECTION]: Ingestion points: Files in world-code/ (voice.md, climax.md, method.md, creation.md, conversation.md, crossing.md) and references/ (ref-open-loop-mechanics.md, ref-sos-world-code-mapping.md, ref-email-copy-guidelines.md, ref-email-types.md).
- [PROMPT_INJECTION]: Boundary markers: Absent. The skill does not use delimiters to isolate external content or instructions.
- [PROMPT_INJECTION]: Capability inventory: Limited to text generation for emails; no file-system writing or network access identified.
- [PROMPT_INJECTION]: Sanitization: Absent. Content from ingested files is used directly without validation.
Audit Metadata