confluence-automation

Warn

Audited by Gen Agent Trust Hub on Mar 5, 2026

Risk Level: MEDIUMEXTERNAL_DOWNLOADSPROMPT_INJECTION
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill requires connecting to an external MCP server endpoint at https://rube.app/mcp. This involves fetching tool definitions and executing logic hosted on a remote server that is not part of the established trusted vendor or well-known service lists.
  • [PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection through the ingestion of external data from Confluence pages.
  • Ingestion points: Data enters the agent context through tools like CONFLUENCE_SEARCH_CONTENT, CONFLUENCE_GET_PAGE_BY_ID, and CONFLUENCE_CQL_SEARCH as specified in SKILL.md.
  • Boundary markers: No specific boundary markers or instructions to ignore embedded commands are implemented to protect the agent from malicious instructions within Confluence content.
  • Capability inventory: The agent is granted significant permissions to modify the environment, including CONFLUENCE_CREATE_PAGE, CONFLUENCE_UPDATE_PAGE, and CONFLUENCE_DELETE_PAGE (SKILL.md).
  • Sanitization: There is no evidence of sanitization or content validation for data retrieved from the remote source before it is processed by the agent.
Audit Metadata
Risk Level
MEDIUM
Analyzed
Mar 5, 2026, 07:55 AM