work-item-sequencing
Pass
Audited by Gen Agent Trust Hub on Sep 2, 2026
Risk Level: SAFENO_CODEINDIRECT_PROMPT_INJECTION
Full Analysis
- [NO_CODE]: The skill consists of markdown instructions and platform metadata without any scripts or executable files.
- [INDIRECT_PROMPT_INJECTION]: The skill establishes a process for interpreting external work item subjects, creating a potential surface for indirect prompt injection. 1. Ingestion points: Subject__c field in GUS work items (referenced in SKILL.md). 2. Boundary markers: Absent. 3. Capability inventory: Logic involves reading and ordering items using tools like gus-cli (referenced in SKILL.md). 4. Sanitization: Absent.
- [SAFE]: The logic for sequencing work items follows standard organizational patterns. No unauthorized data exfiltration, obfuscation, or malicious command execution was found.
Audit Metadata