The Agent Skills Directory

[PROMPT_INJECTION]: The skill provides legitimate instructions for improving test reliability. No patterns targeting agent behavioral overrides or safety filter bypasses were found.
[DATA_EXFILTRATION]: No network operations, hardcoded credentials, or access to sensitive local file paths (e.g., SSH keys, AWS credentials) were detected.
[OBFUSCATION]: The skill's content and code in example.ts are written in plain, human-readable text. No Base64, zero-width characters, Unicode tag characters, or homoglyphs were found.
[REMOTE_CODE_EXECUTION]: The skill does not perform any remote package installations or script executions via curl, wget, or similar tools.
[DYNAMIC_EXECUTION]: The provided TypeScript utilities in example.ts use standard setTimeout polling for condition checking. No use of eval(), exec(), or dynamic runtime code generation from untrusted sources was found.
[INDIRECT_PROMPT_INJECTION]: While the waitForEventMatch function in example.ts processes event data, it is designed for logical comparison during test execution and does not interpolate data into agent prompts or instruction streams.
[DYNAMIC_CONTEXT_INJECTION]: The skill does not use the !command syntax for executing shell commands at skill load time.

condition-based-waiting