skills/openai/plugins/triage-issue/Gen Agent Trust Hub

triage-issue

Pass

Audited by Gen Agent Trust Hub on May 9, 2026

Risk Level: SAFE
Full Analysis
  • Jira API Integration: The skill utilizes specific tools to search, retrieve, and update Jira issues (e.g., searchJiraIssuesUsingJql, createJiraIssue). These operations are within the expected scope for a triage tool and are used to interact with the user's Jira environment.
  • External Data Processing (Indirect Injection Surface): The skill is designed to analyze untrusted data in the form of bug reports and error messages (SKILL.md, Step 1). While this introduces a surface for indirect instructions, the skill mitigates this through several design choices:
  • Human-in-the-Loop: Step 4 explicitly mandates that the agent must present findings and wait for a user decision before taking any action.
  • Boundary Markers: The skill encourages the use of code blocks and structured headers when formatting bug reports, which helps in delimiting external content.
  • Capabilities: The tools used are limited to Jira operations, preventing access to broader system permissions or sensitive local files.
  • Data Handling and Templates: The skill provides detailed templates in references/bug-report-templates.md to ensure that collected information (like error signatures and environment details) is processed in a consistent, structured format, which reduces the risk associated with handling malformed inputs.
Audit Metadata
Risk Level
SAFE
Analyzed
May 9, 2026, 03:11 PM