designing-surveys

Pass

Audited by Gen Agent Trust Hub on Feb 17, 2026

Risk Level: SAFE
Full Analysis
  • [NO_CODE] (SAFE): This skill contains no executable scripts (Python, JavaScript, etc.). It is composed of Markdown documentation and a JSON metadata file. All logic is provided via natural language instructions for the LLM.
  • [DATA_EXFILTRATION] (SAFE): No patterns for file system access or network operations (curl, wget, fetch) were detected. The skill operates entirely within the agent's chat context.
  • [CREDENTIALS_UNSAFE] (SAFE): No hardcoded API keys, tokens, or secrets were found in the provided files.
  • [REMOTE_CODE_EXECUTION] (SAFE): There are no package managers (npm, pip) or remote script downloads present in the skill's definition.
  • [PROMPT_INJECTION] (SAFE): The instructions do not contain attempts to bypass safety filters, extract system prompts, or override agent behavior. The language is professional and task-oriented.
  • [INDIRECT_PROMPT_INJECTION] (SAFE): While the skill ingests user input regarding survey goals, it lacks any high-risk capabilities (like shell access or network requests) that could be exploited via malicious data injection. The output is limited to generating Markdown survey plans.
Audit Metadata
Risk Level
SAFE
Analyzed
Feb 17, 2026, 06:26 PM