evolving-design-system-components

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

• Third-party content exposure detected (high risk: 0.85). The required runtime workflow calls get_confluence_page to read Bitwarden governance pages from Atlassian Confluence; those pages are outsider-authored content (not authored by the operating user) and their body text can be ingested into the agent’s LLM context.

MEDIUM W012: Unverifiable external dependency detected (runtime URL that controls agent).

• Potentially malicious external URL detected (high risk: 0.90). The skill explicitly requires reading external Confluence governance pages at runtime via get_confluence_page (e.g. https://bitwarden.atlassian.net/wiki/spaces/PROD/pages/665780251 and https://bitwarden.atlassian.net/wiki/spaces/PROD/pages/1804206168), and that fetched content is used as the authoritative guidance that directly controls how the agent composes proposals, so this is a runtime external dependency that controls agent instructions.