writing-fragments
Pass
Audited by Gen Agent Trust Hub on Sep 23, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [SAFE]: The skill is designed for creative writing and local file management. It does not require network access, high-privilege commands, or external dependencies.
- [INDIRECT_PROMPT_INJECTION]: The skill involves reading and writing user-provided content to a local Markdown file.
- Ingestion points: The skill captures fragments from user dialogue and re-reads the Markdown file from disk to preserve manual edits (referenced in
SKILL.md). - Boundary markers: There are no explicit instructions to the model to ignore embedded instructions within the text fragments it processes.
- Capability inventory: The skill is capable of writing and appending to local files.
- Sanitization: No explicit sanitization of the user-provided text is performed before it is written to the file.
- Assessment: While this creates a theoretical surface for indirect prompt injection if the resulting file is later used to drive agent behavior, it is a standard characteristic of document-editing tools and does not constitute a malicious finding.
Audit Metadata