skills/jeremylongshore/claude-code-plugins-plus-skills/connection-pooling-config/Gen Agent Trust Hub
connection-pooling-config
Pass
Audited by Gen Agent Trust Hub on Feb 16, 2026
Risk Level: LOWNO_CODE
Full Analysis
- NO_CODE (SAFE): The analyzed file is a markdown document containing only metadata and instructional text. No executable scripts, binaries, or configuration logic are present.
- COMMAND_EXECUTION (LOW): The skill requests access to powerful tools like
BashandWrite. However, there are no predefined commands or scripts within the skill that utilize these tools maliciously. The risk is limited to the agent's general use of these tools based on user prompts. - EXTERNAL_DOWNLOADS (SAFE): No remote scripts, package managers (npm/pip), or external URLs were detected.
- PROMPT_INJECTION (SAFE): The instructions are standard for an AI skill and do not contain override commands, role-play jailbreaks, or instructions to ignore safety filters.
Audit Metadata