The Agent Skills Directory

[EXTERNAL_DOWNLOADS]: The skill downloads an installation script from the vendor's official domain at https://getbruin.com/install/cli during the initial setup phase.
[REMOTE_CODE_EXECUTION]: The skill executes the downloaded CLI installation script directly using the shell (sh). While this originates from the official vendor site, piped remote execution is a significant security pattern.
[COMMAND_EXECUTION]: The agent is instructed to execute various local commands, including git, bruin CLI operations, and MCP configuration commands, which are necessary for setting up the data pipeline.
[INDIRECT_PROMPT_INJECTION]: The skill ingests and processes untrusted data from external sources like Google Sheets and Notion, creating a surface for potential indirect prompt injection attacks.
Ingestion points: Data is pulled from external Sheets and Notion databases as defined in SKILL.md.
Boundary markers: The skill does not define specific delimiters or instructions to ignore potential commands embedded within the data.
Capability inventory: The skill possesses the ability to execute shell commands via the bruin CLI and git.
Sanitization: There is no evidence of sanitization or validation of the data content before it is processed.

sheets-and-notion