new-issue
Pass
Audited by Gen Agent Trust Hub on Jul 28, 2026
Risk Level: SAFEPROMPT_INJECTION
Full Analysis
- [SAFE]: The skill instructions describe a legitimate administrative process for issue management. No malicious patterns, such as prompt injection, obfuscation, or unauthorized system access, were detected.
- [PROMPT_INJECTION]: The skill processes untrusted user input to draft issues, creating an indirect injection surface. Ingestion points: Problem descriptions, desired outcomes, and acceptance criteria are gathered from users (SKILL.md). Boundary markers: The workflow mandates the use of canonical issue structures and refers to external publishing guardrails (SKILL.md). Capability inventory: The skill can publish issues to a remote tracker via an adapter and apply relationships (SKILL.md). Sanitization: The process explicitly requires applying a confidentiality guard and validating all remote values before any mutation is performed (SKILL.md).
Audit Metadata