umb-cms-dev-cli

Pass

Audited by Gen Agent Trust Hub on Sep 26, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTIONEXTERNAL_DOWNLOADSCOMMAND_EXECUTION
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill retrieves content from external Umbraco CMS instances (such as documents and media), which presents a surface for indirect prompt injection. This risk is mitigated by the SDK's built-in validation, which rejects control characters, path traversal sequences, and malformed inputs.
  • [EXTERNAL_DOWNLOADS]: The agent is instructed to use npx @umbraco-cms/mcp-dev@latest to run the tool. This utilizes the official NPM registry to fetch the vendor-provided package, which is a standard and legitimate distribution method.
  • [COMMAND_EXECUTION]: The skill utilizes local shell commands (e.g., ls and node) for environment detection and tool execution. These operations are restricted to the local development context and the specific Umbraco CLI tool.
  • [SAFE]: A detailed analysis of the skill's instructions, metadata, and references revealed no evidence of malicious obfuscation, persistence mechanisms, or unauthorized data exfiltration.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 26, 2026, 07:35 AM