prompt-improver
Pass
Audited by Gen Agent Trust Hub on Mar 21, 2026
Risk Level: SAFENO_CODE
Full Analysis
- [NO_CODE]: The skill consists exclusively of Markdown files (.md) and images. It does not include any JavaScript, Python, shell scripts, or other executable formats, eliminating the risk of direct code execution by the skill itself.
- [EXTERNAL_DOWNLOADS]: No instructions for downloading external packages or executing remote scripts (e.g., via curl or wget) were found. The installation instructions in the README are standard manual steps for the user.
- [COMMAND_EXECUTION]: The skill does not utilize tools to execute system commands. While terminal commands are mentioned in the documentation and prompt examples (e.g., 'root access', 'bash'), these are used as descriptive context within prompt templates rather than commands for the agent to run.
- [DATA_EXFILTRATION]: There are no network-capable commands or references to non-whitelisted domains. The skill does not access sensitive files or environment variables.
- [PROMPT_INJECTION]: The skill's instructions focus on improving the clarity, specificity, and reasoning structure of AI prompts. No bypass patterns, safety-override instructions, or system-prompt extraction attempts were detected.
Audit Metadata