oh-pdd-design-doc-generator
Pass
Audited by Gen Agent Trust Hub on Sep 16, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes user-provided PRD documents as a primary source for generating design content. These documents are untrusted external inputs that could contain hidden instructions designed to override the agent's behavior.
- Ingestion points: PRD document content is read and analyzed as described in
SKILL.md. - Boundary markers: The skill does not provide specific instructions to the agent to treat PRD content strictly as data or use delimiters to prevent instruction injection.
- Capability inventory: The skill utilizes powerful discovery and extraction tools including
Glob,Grep,Read, and theTaskagent exploration tool. - Sanitization: There is no mention of sanitizing or validating the PRD content before it is processed by the agent's reasoning engine.
- [COMMAND_EXECUTION]: The skill instructions direct the agent to use
GlobandGrepto search the codebase for sensitive configuration files (bundle.json,BUILD.gn,sa_profile/*.json) and source code. While these actions are intended for architectural analysis, they provide significant visibility into the local environment that could be exploited if the agent follows malicious instructions from a PRD.
Audit Metadata