The Agent Skills Directory

[REMOTE_CODE_EXECUTION]: The skill instructs the agent to fetch and execute installation scripts from official developer tool repositories.
Evidence: curl -fsSL https://deb.nodesource.com/setup_22.x | bash - in SKILL.md for Node.js installation.
Evidence: curl -sS https://getcomposer.org/installer | php in SKILL.md for Composer installation.
[COMMAND_EXECUTION]: The skill uses standard shell commands to manage system packages and build application assets within a CI/CD environment.
Evidence: apt-get install, docker-php-ext-install, and npm run build in SKILL.md.
[EXTERNAL_DOWNLOADS]: Downloads configuration and installation files from reputable, well-known services.
Evidence: Fetches Node.js setup from NodeSource and Composer installer from getcomposer.org.
[INDIRECT_PROMPT_INJECTION]: The skill manages pipelines that ingest project metadata files which could potentially contain untrusted content.
Ingestion points: Reads composer.json, package.json, and lock files in SKILL.md.
Boundary markers: Absent; files are processed as standard build inputs.
Capability inventory: Includes Bash for command execution and Read for file access.
Sanitization: None; standard behavior for CI/CD pipeline automation.

CI/CD Pipeline Management