jira-create-epic-from-plan

Pass

Audited by Gen Agent Trust Hub on May 16, 2026

Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill uses the Bash tool to perform chmod +x on a dynamically generated shell script (scripts/jira-link-{feature}.sh). This is part of the skill's core functionality to allow the user to manually execute the script for linking Jira issues.
  • [PROMPT_INJECTION]: The skill exhibits an indirect prompt injection attack surface as it reads and parses untrusted external data (Markdown plans or OpenSpec proposals) to generate Jira tickets and executable scripts.
  • Ingestion points: Step 1 reads content from a file path or inline description provided by the user.
  • Boundary markers: The skill lacks explicit delimiters or instructions to ignore embedded commands within the plan documents.
  • Capability inventory: The skill has the capability to write local files, modify file permissions (chmod), and create issues in the vendor's Jira instance via MCP tools.
  • Sanitization: There is no evidence of sanitization or strict schema validation for the data parsed from the plan files.
  • Mitigation: The risk is significantly mitigated by a mandatory human-in-the-loop 'Approval Gate' (Step 5), which requires the user to review a preview of all generated content before any Jira resources are created.
Audit Metadata
Risk Level
SAFE
Analyzed
May 16, 2026, 05:39 AM
Security Audit — agent-trust-hub — jira-create-epic-from-plan