book-target-reader
Pass
Audited by Gen Agent Trust Hub on Jun 22, 2026
Risk Level: SAFE
Full Analysis
- [EXTERNAL_DOWNLOADS]: The skill provides URLs for South Korean book industry data sources, including the Publication Industry Promotion Agency (KPIPA), Kyobo Book Centre, and Aladin. These are reputable services used as data context for the persona generation task and do not involve executable code or sensitive data transfers.
- [PROMPT_INJECTION]: The workflow involves taking input from a preceding skill or user input (1-line persona) and expanding it. While the instructions do not include explicit delimiter markers (e.g., XML tags) for the input data, the skill's purpose is purely generative within a structured framework and does not attempt to bypass LLM safety protocols.
- [COMMAND_EXECUTION]: No shell commands, script executions, or dynamic context injections (
!command) were found in the skill or configuration files. - [CREDENTIALS_UNSAFE]: No API keys, tokens, or hardcoded secrets were detected. The skill correctly identifies that it operates based on internal frameworks or public statistics.
Audit Metadata