The Agent Skills Directory

[EXTERNAL_DOWNLOADS]: The skill facilitates the installation of the @cargo-ai/cli Node.js package. This is an official resource provided by the vendor, getcargohq, and is necessary for the platform's CLI functionality.
[COMMAND_EXECUTION]: The instructions include several shell commands for environment setup and authentication, such as npm install -g @cargo-ai/cli, cargo-ai login, and claude mcp add. These commands are standard for configuring the Model Context Protocol (MCP) server integration.
[INDIRECT_PROMPT_INJECTION]: The skill establishes a vulnerability surface for indirect prompt injection because it processes user-provided data for execution via remote actions.
Ingestion points: The skill ingests untrusted data from the conversation via the data and records arguments in the execute_action and execute_action_batch tools.
Boundary markers: The instructions do not define explicit delimiters or 'ignore' directives to separate user-provided content from the agent's internal tool-calling logic.
Capability inventory: The skill provides access to discover and run various actions (connectors, native operations, and workflows) through a remote MCP server.
Sanitization: No specific input sanitization or data validation methods are described for the content being sent to the remote execution engine.

cargo-mcp