backlog
Pass
Audited by Gen Agent Trust Hub on Sep 30, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill describes operations that ingest and process backlog items which could potentially contain untrusted data.
- Ingestion points: Data returned from
doctrine backlog listanddoctrine backlog show <ID>. - Boundary markers: Not specified in the skill instructions.
- Capability inventory: Ability to create and modify items via
doctrine backlog new,edit, andtagcommands. - Sanitization: No sanitization steps for backlog content are mentioned.
- [SAFE]: No evidence of malicious code, obfuscation, unauthorized network exfiltration, or credential exposure was found. The instructions are purely documentation for standard developer CLI workflows.
Audit Metadata