jira-issue-writing
Pass
Audited by Gen Agent Trust Hub on Oct 7, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [SAFE]: No malicious patterns, obfuscation, or unauthorized execution capabilities were detected. The skill uses natural language to guide the agent through a standard professional workflow.
- [INDIRECT_PROMPT_INJECTION]: The skill has a surface for indirect injection by reading existing Jira issues to check for duplicates. 1. Ingestion points: Existing issues retrieved from Jira projects and boards (SKILL.md, Step 2). 2. Boundary markers: None explicitly defined in the markdown templates, but a manual review step is required before creation (SKILL.md, Step 7). 3. Capability inventory: Searching Jira projects and creating new Jira issues. 4. Sanitization: The skill explicitly mandates redacting secrets, tokens, and private data from any created content.
Audit Metadata