write-script-postgresql
Pass
Audited by Gen Agent Trust Hub on Oct 14, 2026
Risk Level: SAFECOMMAND_EXECUTIONDYNAMIC_EXECUTIONEXTERNAL_DOWNLOADSINDIRECT_PROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill instructs the agent to execute various
wmillCLI commands, such aswmill script preview,wmill script run, andwmill sync push, to interact with the Windmill platform environment. - [DYNAMIC_EXECUTION]: The skill facilitates the creation and execution of dynamic scripts (SQL, TypeScript, Python) within the Windmill runtime. It provides specific instructions on how to structure these scripts and pass arguments to them.
- [EXTERNAL_DOWNLOADS]: The
wmill generate-metadatacommand is used to resolve and update script dependencies. This process may involve fetching packages from external registries, which is a standard part of the script's dependency management lifecycle. - [INDIRECT_PROMPT_INJECTION]: As the skill involves generating and running code based on user intent, there is an inherent surface for indirect prompt injection if user-provided content contains malicious logic. The skill mitigates this by instructing the agent to use proper PostgreSQL parameterization (e.g.,
$1::TEXT) rather than string interpolation.
Audit Metadata