The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill instructs the agent to read and process configuration files like .op.env and .kamal/secrets which contain secret references. These values are used in shell commands via the Bash tool, creating a surface where malicious content within a secret could influence the agent's behavior. Evidence includes: Ingestion points: .op.env, .kamal/secrets, and project Makefile are used to load external values; Boundary markers: There are no explicit instructions to use delimiters or warnings when processing these secrets; Capability inventory: The skill utilizes Bash for command execution and Write/Edit for file management; Sanitization: No sanitization or validation of the secret values is performed before they are used in command execution.- [COMMAND_EXECUTION]: The skill extensively uses the Bash tool to perform operations like op read, op run, and deployment commands. This behavior is directly aligned with the skill's purpose of automating developer workflows using the 1Password CLI.- [EXTERNAL_DOWNLOADS]: The skill references the 1password/load-secrets-action GitHub Action. This is a reference to an official integration from a well-known service used for managing secrets in CI/CD environments.

onepassword-cli-coder