jira-browser-fetch

Pass

Audited by Gen Agent Trust Hub on May 19, 2026

Risk Level: SAFE
Full Analysis
  • [COMMAND_EXECUTION]: The skill launches browser executables using child_process.spawn to enable remote debugging. This is used for the legitimate purpose of extracting Atlassian session cookies from a local browser instance.
  • [PROMPT_INJECTION]: The skill processes untrusted external data from Jira issues and attachments, posing an indirect prompt injection surface.
  • Ingestion points: Jira REST API, HTML, and XML content fetched in scripts/jira-browser-fetch.js.
  • Boundary markers: Absent; data is archived in raw form.
  • Capability inventory: Spawns browser processes and writes to the local file system.
  • Sanitization: Absent; the tool is designed for raw data archival without modification.
Audit Metadata
Risk Level
SAFE
Analyzed
May 19, 2026, 12:15 AM
Security Audit — agent-trust-hub — jira-browser-fetch