grounded-writing
Pass
Audited by Gen Agent Trust Hub on Sep 25, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONEXTERNAL_DOWNLOADS
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes user-supplied text and external facts gathered through research, which presents an attack surface for instructions embedded in data to influence agent behavior.
- Ingestion points: The skill processes text provided for revision and instructs the agent to search for discoverable public facts (SKILL.md).
- Boundary markers: The instructions scope the writing style to user-published text and advise excluding quoted source text or attributed prose (SKILL.md).
- Capability inventory: No arbitrary command execution, file system writes, or sensitive network exfiltration patterns were identified in the skill files.
- Sanitization: The skill lacks explicit instructions for sanitizing or escaping external research data or user input before processing.
- [EXTERNAL_DOWNLOADS]: The style profile includes references to stylistic examples and editorial guidelines from well-known technical blogs and the author's personal website (references/style-profile.md).
Audit Metadata