The Agent Skills Directory

[EXTERNAL_DOWNLOADS]: The skill installs the official @cargo-ai/cli Node.js package, which is a vendor-owned resource for the Cargo platform.
[COMMAND_EXECUTION]: Shell commands are used via the cargo-ai CLI to manage agent life cycles and configurations. It also provides a curl example for direct interaction with platform API endpoints to configure fields not yet exposed by the CLI.
[INDIRECT_PROMPT_INJECTION]: The skill configures agents that can ingest data from external sources (MCP servers and knowledge files), which is a standard attack surface for indirect prompt injection. 1. Ingestion points: Data enters the agent context via connected MCP clients and knowledge file retrieval (RAG). 2. Boundary markers: The provided instructions do not include specific delimiters or isolation protocols for the ingested content. 3. Capability inventory: Configured agents are capable of performing workspace actions and accessing files or models. 4. Sanitization: No explicit sanitization or validation logic is defined for data retrieved from external sources within the skill instructions.

cargo-ai