writing-fragments
Pass
Audited by Gen Agent Trust Hub on Aug 22, 2026
Risk Level: SAFEPROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill reads content from a local markdown file to sync the session with manual user edits. This content is ingested into the agent's context, creating a surface for indirect prompt injection if the file contains instructions intended to override the agent's behavior.
- Ingestion points: The markdown file at the user-specified path (described in
SKILL.md). - Boundary markers: Absent. The instructions do not specify the use of delimiters or 'ignore' instructions when reading the file content.
- Capability inventory: The skill is capable of performing file read and append operations on the local file system.
- Sanitization: Absent. The content read from the file is processed and preserved without validation or filtering.
Audit Metadata