confluence-automation

Pass

Audited by Gen Agent Trust Hub on Jul 31, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADSPROMPT_INJECTION
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill requires the configuration of a remote Model Context Protocol (MCP) server at https://rube.app/mcp. This external endpoint is responsible for providing the tool definitions and handling the logic for Confluence interactions.
  • [PROMPT_INJECTION]: The skill contains a surface for indirect prompt injection because it reads and processes content from external Confluence pages which may contain adversarial instructions.
  • Ingestion points: Data is ingested from Confluence using tools such as CONFLUENCE_SEARCH_CONTENT, CONFLUENCE_GET_PAGE_BY_ID, and CONFLUENCE_CQL_SEARCH as described in SKILL.md.
  • Boundary markers: The instructions lack explicit boundary markers or delimiters to help the agent distinguish between its instructions and the data retrieved from Confluence.
  • Capability inventory: The skill provides the agent with write and delete capabilities, including CONFLUENCE_CREATE_PAGE, CONFLUENCE_UPDATE_PAGE, and CONFLUENCE_DELETE_PAGE, which could be abused if the agent follows instructions embedded in retrieved pages.
  • Sanitization: There is no evidence of content sanitization or validation for the data retrieved from the external Confluence environment.
Audit Metadata
Risk Level
SAFE
Analyzed
Jul 31, 2026, 04:34 PM
Security Audit — agent-trust-hub — confluence-automation