022-tooling-jira

Pass

Audited by Gen Agent Trust Hub on Apr 7, 2026

Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADS
Full Analysis
  • [COMMAND_EXECUTION]: The skill uses the jira CLI to interact with Jira workspaces. It includes a mandatory interactive check using command -v jira or jira version to verify the tool's availability before any issue-related commands are issued.
  • [EXTERNAL_DOWNLOADS]: Provides user guidance for installing the Jira CLI through trusted and well-known package managers such as Homebrew, APT, and Chocolatey.
  • [SAFE]: Contains strict instructions prohibiting the agent from requesting or displaying Jira API tokens or credentials in the chat, instead directing users to perform local configuration via jira configure.
  • [PROMPT_INJECTION]: The skill processes untrusted data by retrieving issue descriptions and comments from Jira. This represents an indirect prompt injection surface inherent to its function of requirement analysis. (Ingestion points: jira issue view output in references/022-tooling-jira.md; Boundary markers: Absent; Capability inventory: jira issue create, jira issue transition in references/022-tooling-jira.md; Sanitization: Absent).
Audit Metadata
Risk Level
SAFE
Analyzed
Apr 7, 2026, 09:33 PM