acceptance-test-authoring

MEDIUM W011: Third-party content exposure detected (indirect prompt injection risk).

• Third-party content exposure detected (medium risk: 0.30). The skill’s required runtime workflow reads and composes outsider-authored free text from repo-controlled spec sources by extracting Markdown step fences into Gherkin features, via references/javascript/extract-gherkin.cjs/references/python/extract_gherkin.py (reads openspec/**/spec.md and copies fenced ```gherkin step lines) and then runs them through references/javascript/cucumber.cjs/references/python/run_acceptance.py.