The Agent Skills Directory

[PROMPT_INJECTION]: The skill analyzes untrusted repository files to generate deployment plans, which presents an indirect prompt injection surface.
Ingestion points: Processes configuration and manifest files including package.json, docker-compose.yml, Procfile, and ORM/schema configuration files.
Boundary markers: No specific delimiters or "ignore embedded instructions" warnings are used when the agent reads the content of these files.
Capability inventory: The skill is capable of executing npm and opendeploy CLI commands and creating files (e.g., wrapper Dockerfiles) in the project directory.
Sanitization: Content extracted from analyzed repository files is not explicitly sanitized or validated before being incorporated into the deployment planning logic.
[COMMAND_EXECUTION]: Uses global CLI tools such as npm and opendeploy to perform environment preflight checks and execute deployment plan sequences.
[EXTERNAL_DOWNLOADS]: Accesses the npm registry to perform version checks and retrieve metadata for the @opendeploydev/cli package.

opendeploy-monorepo