The Agent Skills Directory

[EXTERNAL_DOWNLOADS]: The skill guides the agent to download the buildin CLI installer from https://cdn.buildin.ai/buildin-cli/install. While the instructions mandate explicit user approval before downloading, this represents an external network dependency.
[REMOTE_CODE_EXECUTION]: The skill provides procedures for installing and updating the CLI tool (buildin update). This involves executing code downloaded from a remote CDN. The skill mitigates this by requiring user approval and suggesting the use of SHA-256 hashes to verify installer integrity before execution.
[INDIRECT_PROMPT_INJECTION]: The skill processes content fetched from external Buildin API endpoints (e.g., through buildin markdown get), creating a potential surface for indirect prompt injection if the ingested data contains malicious instructions.
Ingestion points: API responses from buildin search, buildin page get, and buildin markdown get.
Boundary markers: The instructions do not specify the use of clear delimiters or warnings when the agent processes data retrieved from the API.
Capability inventory: The skill can execute the buildin CLI to perform authenticated operations including file uploads and data modifications.
Sanitization: There is no mention of sanitizing or escaping the content retrieved from the API before it is presented to the agent.

buildin-cli