jira-transitions

Pass

Audited by Gen Agent Trust Hub on Nov 9, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill ingests untrusted data from Jira issue summaries and workflow metadata through the get_issue and getTransitions functions. This creates a potential surface where malicious instructions embedded in a Jira issue could attempt to influence the agent's behavior during workflow operations.
  • Ingestion points: In scripts/workflow-demo.mjs (via getIssueStatus and getTransitions) and scripts/workflow-demo.py (via get_issue and get_transitions), where the agent reads issue fields and transition names from the Jira API.
  • Boundary markers: No specific delimiters or "ignore embedded instructions" markers are used when processing the data received from the Jira API.
  • Capability inventory: The skill uses fetch and urllib.request.urlopen to perform network operations, but these are restricted to the Jira API endpoints defined in the environment configuration. It also performs write operations (issue transitions and comments).
  • Sanitization: The skill relies on standard JSON parsing for API responses and does not implement additional sanitization or filtering of the content retrieved from Jira.
Audit Metadata
Risk Level
SAFE
Analyzed
Nov 9, 2026, 06:29 PM