better-docs
Pass
Audited by Gen Agent Trust Hub on Aug 22, 2026
Risk Level: SAFEPROMPT_INJECTION
Full Analysis
- [PROMPT_INJECTION]: The skill is designed to ingest and process untrusted user-provided document drafts (e.g., PRDs, RFCs) to perform editing and rewriting tasks. While the instructions include boundaries to preserve the author's meaning and avoid inventing requirements, the skill lacks explicit boundary markers or 'ignore' instructions to prevent the agent from obeying malicious instructions embedded within the processed text. This creates a surface for indirect prompt injection, where a document could contain instructions to the agent rather than content for editing.
- Ingestion points: The skill reads external drafts provided by the user in chat or files (SKILL.md).
- Boundary markers: Absent. There are no instructions to the agent to disregard formatting or commands embedded within the draft being edited.
- Capability inventory: The skill is instructed to use environment-native tracked-changes, diff mechanisms, and artifact tools to produce output (SKILL.md).
- Sanitization: Absent. The skill does not specify any sanitization or validation of the input text before processing.
Audit Metadata