confluence-automation

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

• Third-party content exposure detected (high risk: 0.75). SKILL.md describes Confluence workflows that call Composio Confluence tools like CONFLUENCE_GET_PAGE_BY_ID / CONFLUENCE_GET_SPACE_CONTENTS / CONFLUENCE_SEARCH_CONTENT, which at runtime can return outsider-authored Confluence page body text (HTML/XHTML) into the agent’s LLM context.

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

• Potentially malicious external URL detected (high risk: 1.00). The skill instructs adding the MCP server endpoint https://rube.app/mcp and to call RUBE_SEARCH_TOOLS at runtime to retrieve tool schemas, meaning that runtime-fetched content from that URL can directly control agent behavior and is a required dependency.