fusion-issue-author-bug

Pass

Audited by Gen Agent Trust Hub on Mar 29, 2026

Risk Level: SAFE
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill processes external data in the form of bug reports and reproduction steps, which constitutes an indirect prompt injection surface. However, the risk is mitigated by design constraints.\n
  • Ingestion points: SKILL.md (Required inputs: Bug context, Reproduction information).\n
  • Boundary markers: Not explicitly defined for the input data interpolation.\n
  • Capability inventory: The skill drafts content to the local filesystem (.tmp/) but lacks the capability to perform GitHub mutations directly.\n
  • Sanitization: None described for the input fields.\n
  • Risk Mitigation: The skill includes a 'Safety & constraints' section that explicitly prohibits direct mutation, delegating that responsibility to a parent orchestrator (fusion-issue-authoring). This architecture ensures that any potentially injected instructions in a bug report cannot be executed by this skill.
Audit Metadata
Risk Level
SAFE
Analyzed
Mar 29, 2026, 01:17 AM