write-testrail-cases

Pass

Audited by Gen Agent Trust Hub on Jul 27, 2026

Risk Level: SAFEPROMPT_INJECTIONEXTERNAL_DOWNLOADS
Full Analysis
  • [PROMPT_INJECTION]: The skill ingests untrusted data from Jira tickets, which creates a surface for indirect prompt injection attacks.
  • Ingestion points: The agent fetches and processes User Story Summary, Description, and Acceptance Criteria via Jira MCP or manual input (SKILL.md Step 3).
  • Boundary markers: No explicit delimiters or instructions are provided to the agent to treat the Jira content as untrusted or to ignore embedded commands.
  • Capability inventory: The skill is configured to perform POST requests to the TestRail API to add test cases (SKILL.md "Add case request").
  • Sanitization: No explicit sanitization logic is defined for the external Jira content before it is used to generate test case scenarios.
  • Mitigation: The workflow includes a mandatory human review step ('wait for explicit confirmation') before any data is posted to TestRail, reducing the impact of potential injection.
  • [EXTERNAL_DOWNLOADS]: The skill fetches supplemental metadata and translation files from external sources.
  • Source URLs: Official GitHub repositories under the 'folio-org' organization (Author Context).
  • Files fetched: translation files (en_US.json), package metadata (package.json), and Cypress testing fragments (SKILL.md "Lightweight Context Enrichment").
  • Usage: Data is used as context for scenario analysis and UI text verification to improve the accuracy of generated test cases.
Audit Metadata
Risk Level
SAFE
Analyzed
Jul 27, 2026, 11:59 AM
Security Audit — agent-trust-hub — write-testrail-cases