writing-fragments
Pass
Audited by Gen Agent Trust Hub on Jul 9, 2026
Risk Level: SAFEPROMPT_INJECTION
Full Analysis
- [PROMPT_INJECTION]: The skill possesses an indirect prompt injection surface by re-reading the target markdown file from disk before every write operation. This behavior allows the content of the file to enter the agent's context, potentially influencing its subsequent actions or the way it interviews the user.
- Ingestion points: The local markdown file path specified by the user (as described in the 'Writing rhythm' section).
- Boundary markers: The skill uses horizontal rules (
---) to separate fragments, but it does not provide explicit instructions for the agent to ignore or isolate potential instructions embedded within those fragments. - Capability inventory: The skill is capable of reading from and appending to local files.
- Sanitization: There is no evidence of sanitization, validation, or filtering of the content read from the external file.
Audit Metadata