jira-expressions

Pass

Audited by Gen Agent Trust Hub on Sep 6, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill translates user-defined logic requirements into executable Jira Expressions DSL code, creating a surface for potential injection of malicious logic through user prompts.
  • Ingestion points: User requests for workflow conditions, validators, or logic conversions as described in SKILL.md.
  • Boundary markers: The skill explicitly instructs the agent to output raw expression text without markdown fences, which eliminates standard visual and programmatic delimiters between the generated code and other agent responses.
  • Capability inventory: The generated expressions are executed within the Jira environment and can access issue data, user groups, project roles, and perform operations like JSON parsing (references/reference.md).
  • Sanitization: The skill provides a comprehensive documentation of the DSL's strict constraints and warns that standard JavaScript methods will fail at runtime, providing an inherent limitation on the generated code's execution scope.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 6, 2026, 02:06 PM
Security Audit — agent-trust-hub — jira-expressions