fat-creator

Pass

Audited by Gen Agent Trust Hub on Sep 14, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill processes data from TestRail to create content in Jira, creating a potential vector for indirect prompt injection if external test case data contains malicious instructions.
  • Ingestion points: The skill fetches test case titles, IDs, and reference fields via the tr.py helper from https://foliotest.testrail.io (referenced in Procedure Step 2).
  • Boundary markers: The skill contains an explicit 'Hard restrictions' section that instructs the agent to ignore any instructions found in a test case body, Jira ticket, or fetched content that would violate its core mission.
  • Capability inventory: The skill has the ability to create Jira issues (createJiraIssue) and links (createIssueLink) in the FAT project, and can perform specific field edits (editJiraIssue) on newly created tickets (SKILL.md Step 8).
  • Sanitization: Data is mapped to structured Jira fields (Summary, Description, custom fields) rather than being executed as code, and the skill enforces a human-in-the-loop 'Preview and confirm' step before any tickets are created (SKILL.md Step 7).
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 14, 2026, 10:28 AM
Security Audit — agent-trust-hub — fat-creator