workflow-create-prd
Pass
Audited by Gen Agent Trust Hub on Sep 28, 2026
Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill instructs the agent to use the Bash tool to perform codebase exploration and technical stack analysis. This is a standard capability for development-oriented agents to understand project structure and existing patterns.
- [INDIRECT_PROMPT_INJECTION]: The skill processes potentially untrusted data from both user responses and existing codebase files without explicit sanitization, creating a surface where embedded instructions could theoretically influence agent behavior. However, this is inherent to the skill's primary purpose of code analysis and requirement gathering.
- Ingestion points: The agent reads user descriptions via 'AskUserQuestion' and explores the codebase using 'Read' and 'Glob'.
- Boundary markers: No specific delimiters or safety warnings are included for the data being processed.
- Capability inventory: The agent uses 'Bash', 'Read', 'Write', and 'Glob' tools.
- Sanitization: No explicit sanitization or validation logic is defined for the input data.
Audit Metadata