The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill includes functionality to analyze existing project files to identify automation opportunities, creating a surface for indirect prompt injection.\n
Ingestion points: The WorkflowAnalyzer script in resources/implementation-playbook.md reads and parses project files including .github/workflows/*.y*ml, .gitlab-ci.yml, and README.md to discover manual processes.\n
Boundary markers: No explicit boundary markers or instructions to ignore embedded instructions are present when the agent processes the discovered content.\n
Capability inventory: The skill facilitates the creation of comprehensive CI/CD pipelines with capabilities for file system access, network operations (e.g., Slack notifications, Docker pushes), and cloud infrastructure management (e.g., AWS ECS, Terraform).\n
Sanitization: The skill does not implement sanitization or filtering of the data extracted from analyzed project files before incorporating it into the workflow analysis.\n- [EXTERNAL_DOWNLOADS]: The implementation playbook includes workflow templates that utilize established and well-known external resources.\n
The templates configure GitHub Actions from official and reputable providers such as GitHub (actions/checkout, actions/setup-node), AWS (aws-actions/configure-aws-credentials), Aqua Security (aquasecurity/trivy-action), and Cypress (cypress-io/github-action).\n
The development setup script downloads standard packages via npm and pip, such as pre-commit, semantic-release, and various linting tools.

cicd-automation-workflow-automate