openspec-new-change
Pass
Audited by Gen Agent Trust Hub on Apr 2, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill provides instructions for orchestrating local development workflows using the
openspecCLI. All operations are local to the user's environment. - [COMMAND_EXECUTION]: The skill executes
openspeccommands using user-provided input for the change name. It includes a specific guardrail requiring the name to be in kebab-case, which serves as a validation step to prevent shell metacharacter injection. - [PROMPT_INJECTION]: No evidence of prompt injection, role-play bypasses, or safety instruction overrides were found. The instructions are task-oriented and respect operational boundaries.
- [DATA_EXFILTRATION]: No network operations, credential access, or sensitive file reads were detected.
Audit Metadata