The Agent Skills Directory

[SAFE]: The skill focuses on legitimate development practices for Ruby and Rails test coverage analysis. All provided scripts, such as 'scripts/collate_coverage.rb' and 'scripts/coverage_analyzer.rb', are utility tools designed to process local coverage data generated by the SimpleCov gem. No malicious patterns or security risks were identified across the 13 analyzed files.\n- [COMMAND_EXECUTION]: The skill includes documentation and scripts that execute standard shell commands like 'bundle exec rake test' and 'git rev-parse'. These are routine operations for a test coverage tool intended for use in developer environments or CI/CD pipelines. Examples of command execution in the documentation are clearly scoped to the tool's primary purpose of running tests or quality analysis.\n- [DATA_EXFILTRATION]: No sensitive data access or exfiltration patterns were detected. An example for a custom Slack notification formatter is provided in 'references/advanced_patterns.md' for educational purposes; it correctly demonstrates the use of environment variables for secret management, which is a standard security practice.\n- [INDIRECT_PROMPT_INJECTION]: The skill processes test result JSON files and RubyCritic reports at runtime. While this involves ingesting external data, the analysis is confined to numerical calculations and static HTML report generation using ERB. There is no capability for this ingested data to influence the AI agent's instructions or execute arbitrary logic.

simplecov