The Agent Skills Directory

[EXTERNAL_DOWNLOADS]: The skill instructs the user to install the @membranehq/cli tool from the official NPM registry. This package is provided by the vendor to enable interaction with their integration platform.
[COMMAND_EXECUTION]: Utilizes the membrane CLI to manage authentication, establish connections to DoltHub, and execute database actions. These commands are consistent with the skill's stated purpose.
[SAFE]: The skill processes data from external actions, which introduces a theoretical surface for indirect prompt injection, although no specific exploits are present.
Ingestion points: Output from commands such as membrane action list and membrane action run in SKILL.md is ingested into the agent's context.
Boundary markers: Not present; the skill does not define specific delimiters for separating tool output from instructions.
Capability inventory: Includes shell command execution via the membrane CLI and package management via npm as documented in SKILL.md.
Sanitization: No explicit sanitization or filtering of external data is described in the skill instructions.

dolthub