The Agent Skills Directory

[EXTERNAL_DOWNLOADS]: The skill recommends installing the @membranehq/cli package via NPM. This is an official tool provided by the platform vendor (Membrane) to manage API connections and is used here for its intended purpose.
[COMMAND_EXECUTION]: The agent is instructed to use various membrane CLI commands, such as login, connect, and action run, to manage authentication and interact with the JP Funda API. These are standard functional commands for the integration.
[PROMPT_INJECTION]: The skill facilitates the ingestion of external data from the JP Funda API through membrane action run and membrane request commands (Ingestion point: SKILL.md). There are no explicit boundary markers or sanitization requirements specified for handling this external content. Since the agent has the capability to execute shell commands via the CLI (Capability inventory: SKILL.md), this creates a surface for indirect prompt injection where malicious instructions embedded in an API response could influence the agent's actions. Additionally, the metadata description ('Manage Organizations, Pipelines, Users, Filters') is inconsistent with the primary financial reporting functionality, which is misleading.

jp-funda