The Agent Skills Directory

[EXTERNAL_DOWNLOADS]: The skill requires the installation of the @membranehq/cli Node.js package.
Evidence: npm install -g @membranehq/cli in SKILL.md.
Context: This is a vendor-owned CLI tool used to manage authentication and execute requests to the Microsoft Outlook API.
[COMMAND_EXECUTION]: The skill uses the membrane CLI to perform operations and manage service connections.
Evidence: Instructions for membrane login, membrane connect, and membrane action run are provided in SKILL.md.
Context: These commands are used to interact with the environment and the vendor's API gateway.
[PROMPT_INJECTION]: The skill has a surface for indirect prompt injection as it ingests untrusted data from an external service.
Ingestion points: Reads email content, calendar events, and contact information from Microsoft Outlook (SKILL.md).
Boundary markers: No explicit delimiters or instructions to ignore embedded commands are specified in the instructions for processing external content.
Capability inventory: The agent can execute actions (membrane action run) and raw API requests (membrane request), allowing for state-changing operations like sending emails or creating events.
Sanitization: No explicit filtering or sanitization of the retrieved data is mentioned in the skill documentation.

microsoft-outlook