jira-cli

Pass

Audited by Gen Agent Trust Hub on Feb 19, 2026

Risk Level: SAFEPROMPT_INJECTIONNO_CODECOMMAND_EXECUTION
Full Analysis
  • [Indirect Prompt Injection] (LOW): The skill provides commands to retrieve data from Jira issues and comments, creating an ingestion surface for untrusted external instructions. * Ingestion points: 'jira issue list', 'jira issue view', and 'jira issue view --comments' as documented in 'SKILL.md'. * Boundary markers: Absent; there are no instructions for the agent to use delimiters or ignore instructions found within the ticket data. * Capability inventory: Command execution via the 'jira' CLI. * Sanitization: Absent; no content validation or sanitization routines are specified.
  • [No Code] (SAFE): The skill package consists solely of markdown documentation and does not include any executable scripts, binaries, or active code components.
  • [Command Execution] (SAFE): The skill documents standard Jira management commands. These commands do not include suspicious patterns such as arbitrary shell piping, obfuscation, or privilege escalation.
Audit Metadata
Risk Level
SAFE
Analyzed
Feb 19, 2026, 05:42 PM