The Agent Skills Directory

[REMOTE_CODE_EXECUTION]: The skill instructs the agent to install the Buildin CLI by downloading a script from https://cdn.buildin.ai/buildin-cli/install and piping it directly into bash. A similar pattern is used for Windows PowerShell using irm ... | iex at https://cdn.buildin.ai/buildin-cli/install.ps1.
[COMMAND_EXECUTION]: The skill's primary function is to execute shell commands using the buildin CLI to interact with the Buildin V2 API, manage files, and configure authentication.
[EXTERNAL_DOWNLOADS]: The skill performs network requests to download the CLI binary, installation scripts, and updates from cdn.buildin.ai.
[PROMPT_INJECTION]: The skill has a surface for indirect prompt injection as it retrieves and processes Markdown and JSON content from external Buildin API endpoints.
Ingestion points: Data enters the context via commands like buildin markdown get, buildin page get, and buildin database query from the api.buildin.ai endpoint.
Boundary markers: No explicit boundary markers or delimiters are defined to isolate the processed external content from the agent's instructions.
Capability inventory: The agent can execute various shell commands through the buildin CLI, including data modification (put), file uploads, and search operations.
Sanitization: The instructions include behavioral guidelines to confirm IDs and file paths against user-authorized input, but there is no technical sanitization of the data retrieved from the API.

buildin-cli