The Agent Skills Directory

[COMMAND_EXECUTION]: The skill utilizes a bundled Python CLI script located at scripts/datalore, which is executed using the uv tool. This CLI serves as the primary interface for managing Datalore notebook cells, executing code, and interacting with attached databases.
[INDIRECT_PROMPT_INJECTION]: The skill processes content from notebook cells, database schemas, and file previews which could contain instructions designed to influence the agent's behavior.
Ingestion points: Content is ingested through commands like scripts/datalore cells --include-outputs, scripts/datalore cell get, scripts/datalore db condensed, and scripts/datalore file text.
Boundary markers: The instructions do not define specific delimiters for separating external data from system instructions, nor do they include warnings to ignore instructions found within data.
Capability inventory: The agent has the ability to create, edit, execute, and delete notebook cells and files, providing a significant surface for actions triggered by injected content.
Sanitization: There is no evidence of content sanitization or filtering applied to data retrieved from the Datalore API before it is processed by the agent.
[SAFE]: The skill uses standard security practices for authentication, including support for the system keychain via the keyring library and OAuth 2.0 PKCE for browser-based authorization. All network communications are directed to the vendor's official domain (datalore.jetbrains.com).

datalore-notebook