document-builder
Pass
Audited by Gen Agent Trust Hub on Jul 11, 2026
Risk Level: SAFE
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill is designed to transform untrusted user input, such as rough notes and briefs, into structured documents. This establishes a standard surface for indirect prompt injection, although the risk is limited to the skill's text transformation role.
- Ingestion points: User-provided notes, conversations, and source materials (SKILL.md).
- Boundary markers: Not explicitly defined in the instructions to prevent the agent from following instructions embedded in source data.
- Capability inventory: The skill is primarily focused on text generation and file creation.
- Sanitization: No specific sanitization or filtering logic is described.
- [COMMAND_EXECUTION]: The skill includes a reference template for designing other agent skills (references/general/agent-skill-architecture.md) that mentions dynamic context injection using shell commands like 'pwd' or 'cat package.json'. These are presented as educational examples for developers building new skills and do not represent an active vulnerability within this documentation skill.
Audit Metadata