The Agent Skills Directory

[COMMAND_EXECUTION]: The skill relies on executing shell commands via the @useclawlink/cli. This is the primary mechanism for the agent to interact with the Replicate service.
[EXTERNAL_DOWNLOADS]: The skill uses npx to fetch and execute the @useclawlink/cli package from the NPM registry. This is a standard delivery method for the vendor's management tools.
[INDIRECT_PROMPT_INJECTION]: The skill passes user instructions directly into command-line arguments, which could be exploited to influence the behavior of the CLI tool.
Ingestion points: User-provided strings are interpolated into the <what you want to do> argument for the npx @useclawlink/cli actions replicate command in SKILL.md.
Boundary markers: None present; user input is enclosed in double quotes within the shell command.
Capability inventory: The skill has the capability to create and cancel predictions, create models and deployments, and upload files to Replicate via the CLI tool (SKILL.md).
Sanitization: There is no explicit sanitization or validation of the user-provided string before it is passed to the shell command in the provided instructions.

replicate