monday-automation

Pass

Audited by Gen Agent Trust Hub on Mar 5, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: The skill is a configuration-based extension that uses instructions rather than executable code to facilitate Monday.com operations through a third-party MCP provider.
  • [EXTERNAL_DOWNLOADS]: The skill references a remote MCP server endpoint (https://rube.app/mcp) which is the standard service delivery method for Rube/Composio tools. This reference is documented neutrally as it is essential for the skill's stated functionality.
  • [SAFE]: Authentication is handled via the RUBE_MANAGE_CONNECTIONS workflow, which uses Monday.com OAuth rather than requiring hardcoded API keys or secrets within the skill itself.
  • [PROMPT_INJECTION]: A theoretical indirect prompt injection surface exists because the skill reads content from Monday.com boards (external data) and possesses write capabilities (item updates, GraphQL mutations).
  • Ingestion points: MONDAY_LIST_BOARD_ITEMS, MONDAY_ITEMS_PAGE, MONDAY_LIST_SUBITEMS_BY_PARENT in SKILL.md
  • Boundary markers: Not explicitly defined in the prompt templates.
  • Capability inventory: Includes item creation, column updates, and raw GraphQL mutations via MONDAY_CREATE_OBJECT.
  • Sanitization: Not explicitly implemented in the instruction text. This is considered a low-risk inherent characteristic of automation tools rather than a malicious defect.
Audit Metadata
Risk Level
SAFE
Analyzed
Mar 5, 2026, 07:56 AM