The Agent Skills Directory

[DYNAMIC_CONTEXT_INJECTION]: The skill uses dynamic context commands (using the ! syntax) in the SKILL.md file to detect the project environment at load time. These commands are limited to find and grep operations in the root directory to identify standard configuration files like package.json, pyproject.toml, or playwright.config.ts. These are benign utility operations for project discovery.
[COMMAND_EXECUTION]: The skill generates shell commands to execute tests based on user-provided file paths. While it uses variable interpolation (e.g., "$FILE"), it correctly wraps variables in double quotes to prevent basic shell splitting. The tools utilized (bunx, python, cargo, go) are standard for software development workflows.
[SAFE]: No malicious patterns, data exfiltration, obfuscation, or persistence mechanisms were detected. The skill's behavior matches its described purpose of optimizing local test iteration.

test-focus