write-script-postgresql

Pass

Audited by Gen Agent Trust Hub on Aug 18, 2026

Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill instructs the agent to execute shell commands via the wmill CLI, including wmill script preview, wmill script run, and wmill generate-metadata. These commands are used to test and manage scripts within the Windmill developer workflow.
  • [INDIRECT_PROMPT_INJECTION]: The skill accepts user-supplied arguments that are interpolated into shell commands for script execution (e.g., wmill script preview <path> -d '<args>').
  • Ingestion points: User-supplied script arguments are ingested through command-line flags during the preview process.
  • Boundary markers: The command template uses single quotes to wrap user arguments.
  • Capability inventory: The agent has the capability to execute shell commands and trigger the execution of database logic with potential side effects.
  • Sanitization: The instructions do not specify any mandatory sanitization or validation of the user-provided arguments.
  • [UNVERIFIABLE_DEPENDENCIES_AND_REMOTE_CODE_EXECUTION]: The wmill generate-metadata command resolves and updates project dependencies. The instructions mitigate risk by requiring the agent to notify the user of any dependency version changes detected during this operation.
Audit Metadata
Risk Level
SAFE
Analyzed
Aug 18, 2026, 04:44 AM
Security Audit — agent-trust-hub — write-script-postgresql