jira-cli
Pass
Audited by Gen Agent Trust Hub on Sep 15, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill exposes an indirect prompt injection attack surface by fetching untrusted external content from Jira tickets that could influence agent behavior.
- Ingestion points: External data enters the agent context in
SKILL.mdthrough commands that retrieve issue data, such asjira issue view,jira issue list, and viewing comments. - Boundary markers: Absent. There are no specific delimiters or instructional guardrails to prevent the agent from executing commands or text embedded within the retrieved issue summaries or bodies.
- Capability inventory: The skill enables full interaction with the system shell via the
jirabinary inSKILL.mdto create, modify, transition, and delete issues. - Sanitization: Absent. No escaping or structural validation is provided for incoming data before it is processed by the model.
Audit Metadata