prd-starter-configurator
Pass
Audited by Gen Agent Trust Hub on Mar 24, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill serves as a structured questionnaire to facilitate project setup. It does not include executable scripts, shell commands, or external dependencies. All operations are confined to user interaction and data modeling.- [PROMPT_INJECTION]: The skill incorporates untrusted user input (such as tech stack descriptions and project names) into a final JSON structure. This establishes an indirect prompt injection surface where maliciously crafted user input could potentially influence the behavior of agents that consume the resulting configuration.
- Ingestion points: Phase 3.1 (name, description) and Phase 3.3 (tech stack) in
SKILL.md. - Boundary markers: The skill applies a regex check (
^[a-z0-9-]+$) to the project name, but descriptive fields (description, tech stack) are ingested as free-form text without specific boundary markers or 'ignore' instructions. - Capability inventory: The agents configured by this skill are granted access to sensitive MCP servers including
github,filesystem,gitkraken, andplaywright. - Sanitization: There are no instructions for sanitizing or escaping the user-provided strings before they are interpolated into the JSON output.
Audit Metadata