writing-fragments
Pass
Audited by Gen Agent Trust Hub on Jul 27, 2026
Risk Level: SAFEPROMPT_INJECTION
Full Analysis
- [PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection. 1. Ingestion points: The agent is instructed to re-read the target markdown file from disk during every turn to preserve user changes (SKILL.md). 2. Boundary markers: The instructions do not provide delimiters or warnings to treat the file content as data rather than instructions. 3. Capability inventory: The skill utilizes file-writing capabilities to append or edit fragments. 4. Sanitization: No content filtering or validation is performed on the data retrieved from the file. This lack of isolation could allow instructions within a writing fragment to influence the agent's behavior.
Audit Metadata