jira-expert
Pass
Audited by Gen Agent Trust Hub on Sep 4, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill interacts with external Jira data which may contain untrusted content (such as issue descriptions or comments), potentially influencing agent behavior. This is a standard characteristic of project management skills.
- Ingestion points: Reads Jira issue content, comments, and project configurations via MCP tools (SKILL.md).
- Boundary markers: The skill uses structured tool parameters for interaction, providing implicit segmentation between data and instructions.
- Capability inventory: Includes capabilities for project modification, issue creation, and running local utility scripts (scripts/jql_query_builder.py, scripts/workflow_validator.py).
- Sanitization: The skill relies on the underlying Jira API and MCP server protocols for data validation.
- [SAFE]: The provided Python utilities strictly use standard libraries for logic and regex-based parsing. No evidence of obfuscation, remote code execution, persistence mechanisms, or unauthorized data exfiltration was found.
Audit Metadata