writing-fragments
Pass
Audited by Gen Agent Trust Hub on Oct 9, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill interacts with user-controlled markdown files, creating a potential surface for indirect prompt injection. It is instructed to read the file from disk before each write operation to preserve user edits, which involves the agent processing the file's content.
- Ingestion points: The skill ingests text directly from user prompts and reads back content from a markdown file located at a user-specified path (SKILL.md).
- Boundary markers: There are no explicit instructions to use delimiters or ignore instructions that may be embedded within the fragments during the read-back phase.
- Capability inventory: The skill performs file-read and file-write (append) operations based on user instructions.
- Sanitization: No specific sanitization or filtering logic is implemented for the content read from the disk or received from the user.
Audit Metadata