gsd-new-project
Pass
Audited by Gen Agent Trust Hub on Jul 11, 2026
Risk Level: SAFEPROMPT_INJECTION
Full Analysis
- [SAFE]: The skill provides a project initialization utility that generates planning documents from local templates. No malicious code, obfuscation, or unauthorized access patterns were detected.
- [PROMPT_INJECTION]: The skill processes user-provided project descriptions to initialize context, creating a surface for indirect prompt injection. This is a common architectural pattern for context-gathering tools. \n
- Ingestion points: Processes an idea document provided by the user via @ reference. \n
- Boundary markers: No explicit markers are defined in the skill wrapper to isolate untrusted input. \n
- Capability inventory: The workflow utilizes
Bash,Write, andAgenttools across the project structure. \n - Sanitization: Input is incorporated into planning documents without specific validation filters mentioned in the skill definition.
Audit Metadata