The Agent Skills Directory

[SAFE]: The skill instructions and metadata contain no malicious intent, obfuscation, or data exfiltration patterns. The instructions are focused on legitimate repository maintenance tasks.
[COMMAND_EXECUTION]: The skill defines a specific, hardcoded shell command for running Biome: biome check --write --unsafe --changed --no-errors-on-unmatched --files-ignore-unknown=true --reporter=json. This command is constrained to local operations on modified files and uses structured JSON output to ensure the agent can safely parse diagnostics during the remediation process.
[SAFE]: The skill implements security best practices by requiring repository-native validation (such as build and test scripts) to pass after any automated code changes are made by the Biome tool.

biome-enforcement