spec-creation
Pass
Audited by Gen Agent Trust Hub on Jul 30, 2026
Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill utilizes standard command-line tools (ls, grep) to search the local file system for existing implementation patterns and specifications. These commands are restricted to read-only discovery within the expected project structure.- [PROMPT_INJECTION]: The skill handles untrusted data in the form of business requirements and user stories to generate implementation specifications. This represents a potential surface for indirect prompt injection.
- Ingestion points: Business requirements and user stories provided to the agent for specification generation (SKILL.md).
- Boundary markers: Absent; the skill does not define specific delimiters or instructions to ignore embedded commands within the input requirements.
- Capability inventory: File system discovery (ls, grep) and file writing capabilities to specific output locations like the specs/ directory.
- Sanitization: None; the skill relies on the agent's inherent safety filters and does not specify additional validation for external content.
Audit Metadata