The Agent Skills Directory

[PROMPT_INJECTION]: The skill has an indirect prompt injection surface as it processes configuration files and environment variables to automate tasks.
Ingestion points: Reads package.json, Makefile, and environment variables in dev-setup.sh and deploy.sh.
Boundary markers: No explicit markers are used to isolate content from these files when processed by the agent.
Capability inventory: The skill uses the Bash tool to execute shell scripts and performs remote commands via SSH in scripts/deploy.sh.
Sanitization: No explicit sanitization of configuration file content is performed prior to execution.
[COMMAND_EXECUTION]: The skill generates and executes shell scripts for development setup and deployment. This includes operations like npm install, docker-compose, and SSH-based remote commands. These actions are standard for the skill's intended purpose of workflow automation.
[EXTERNAL_DOWNLOADS]: The skill uses well-known services and trusted organizations to manage dependencies and CI/CD pipelines. This includes using npm for package installation and official GitHub Actions for automated testing. These downloads are considered safe within the context of development automation.

workflow-automation