The Agent Skills Directory

[SAFE]: The skill instructions and associated files demonstrate standard behavior for a developer-focused test runner and analysis tool. No malicious code, exfiltration attempts, or safety bypasses were detected.\n- [OBFUSCATION]: Base64 encoding is utilized in documentation examples for the analyze and inspect commands to identify code fragments. These strings decode to benign programmatic snippets (e.g., return (celsius * 9) / 5 + 32;) and do not contain hidden URLs.\n- [EXTERNAL_DOWNLOADS]: The skill mentions downloading updates from https://update.wallabyjs.com and installing @wallabyjs/cli via standard package managers. These operations target the vendor's official infrastructure and well-known registries, which is standard behavior for this tool.\n- [INDIRECT_PROMPT_INJECTION]: The skill processes project source code and test output to generate reports. This constitutes a standard vulnerability surface where malicious content in scanned files could theoretically attempt to influence the agent via report content. This is considered low-risk and inherent to the functionality of test-related tooling.

wallaby-cli