jira-ticket-refinement

Pass

Audited by Gen Agent Trust Hub on Sep 27, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTIONNO_CODE
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill defines a surface for processing untrusted content from Jira ticket descriptions.
  • Ingestion points: Data is ingested from external Jira issues via the getJiraIssue tool.
  • Boundary markers: The instructions do not specify the use of delimiters or 'ignore embedded instructions' prompts when handling ticket content.
  • Capability inventory: The skill can modify tickets using editJiraIssue and is instructed to read local codebase context.
  • Sanitization: There is no evidence of explicit validation or sanitization for the content processed from Jira.
  • [SAFE]: All identified external resources, such as the atlassian.net domain and the Cloud ID, belong to the vendor's own infrastructure (factorial-io). The use of identifiers for resource discovery is a standard part of the MCP Atlassian toolset.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 27, 2026, 01:04 PM