write-validation
Pass
Audited by Gen Agent Trust Hub on Nov 15, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTIONEXTERNAL_DOWNLOADSDYNAMIC_EXECUTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill documents the
promptmethod, which allows for LLM-based semantic validation of data. This creates a surface for indirect prompt injection if the data being validated contains malicious instructions. 1. Ingestion points: Data tables passed topb.Validate(data=df)inSKILL.md. 2. Boundary markers: The documentation does not specify the use of delimiters or instructions to ignore embedded commands. 3. Capability inventory: Validation failures can triggerActionssuch as Slack notifications or custom Python functions. 4. Sanitization: No sanitization methods for input data are described for thepromptcheck. - [COMMAND_EXECUTION]: The skill documentation includes examples of using
ActionsandFinalActionsto execute Python callables (e.g.,sys.exit(1),pb.send_slack_notification) in response to validation results, which is a standard feature for automated alerting and pipeline orchestration. - [EXTERNAL_DOWNLOADS]: The skill requires the
pointblankPython package, as specified in the compatibility metadata inSKILL.md. - [DYNAMIC_EXECUTION]: The skill supports dynamic logic through the
preargument (e.g., lambda functions for data filtering) and thespeciallymethod, which allows users to define and execute custom validation logic at runtime.
Audit Metadata