response-length-calibration

Pass

Audited by Gen Agent Trust Hub on Apr 3, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: The skill provides best practices and logic for calibrating message lengths for sales bots. Analysis of the instructions and code snippets found no security violations.\n- [COMMAND_EXECUTION]: Python code blocks are included for illustrative logic (length checking, template selection). These scripts do not utilize functions for executing shell commands or interacting with the host system's OS layer.\n- [DATA_EXFILTRATION]: The skill does not contain any network-facing code or patterns that would lead to the unauthorized transmission of data.\n- [PROMPT_INJECTION]: The instructions are strictly limited to the intended domain of message calibration. There are no attempts to override system prompts or bypass safety guardrails.
Audit Metadata
Risk Level
SAFE
Analyzed
Apr 3, 2026, 05:46 PM