drawio-databricks

Pass

Audited by Gen Agent Trust Hub on Sep 10, 2026

Risk Level: SAFEPROMPT_INJECTIONDYNAMIC_EXECUTIONEXTERNAL_DOWNLOADSCOMMAND_EXECUTION
Full Analysis
  • [PROMPT_INJECTION]: The subagent prompt contains a directive to 'Do NOT invoke any drawio skill', which instructs the agent to ignore its usual skill-selection logic in favor of the provided instructions.
  • [DYNAMIC_EXECUTION]: The skill uses a workflow where it scaffolds a JavaScript file (build.mjs) and subsequently executes it using node. This involves runtime execution of agent-modified code generated from templates.
  • [EXTERNAL_DOWNLOADS]: The skill requires the installation of the 'drawio-ai-kit' from GitHub. Although it appropriately instructs the user to perform this installation manually, it introduces a dependency on external code.
  • [COMMAND_EXECUTION]: The instructions rely on the execution of multiple shell commands, such as 'drawio-ai', 'node', and environment variable expansion, to perform its core tasks.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 10, 2026, 01:53 PM
Security Audit — agent-trust-hub — drawio-databricks