The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill requires the agent to read and process external .env.example files or upstream documentation. This represents an attack surface where malicious instructions could be embedded in the data to influence agent behavior.
Ingestion points: The skill workflow involves reading external .env.example files as the primary source of configuration data (SKILL.md).
Boundary markers: No delimiters or specific instructions to ignore embedded commands in the input data are present.
Capability inventory: The skill has the capability to write files (.env, .env.op) and execute complex shell commands (docker compose, op run).
Sanitization: There are no mechanisms described for sanitizing or validating the content of the external configuration files.
[COMMAND_EXECUTION]: The skill relies on the agent executing shell commands to perform infrastructure tasks, secret resolution, and network verification.
Evidence: Instructions include executing op run, docker compose up, openssl rand, and network diagnostic tools like ss and netstat. These commands are consistent with the skill's role as a deployment assistant.

delonet-dotenv