lark-agent-simple

You can create Lark tasks from Markdown test files by running a slash command that parses the document into a compact JSON structure and directly executes Lark MCP calls to build a 3-level task hierarchy with minimal token usage.

The best way to reduce token usage is bypassing heavy workflow generation. This Skill parses Markdown into a minimal data model and directly creates the test, scenario, and task hierarchy via MCP calls to eliminate intermediate processing overhead.

Yes, you can assign owners and due dates when generating Lark tasks from Markdown. The Skill supports optional parameters like --owner, --due-date, and --start-date to configure the directly created 3-level Lark task hierarchy.

Direct MCP execution for Markdown task parsing works by using a slash-command workflow to immediately invoke Lark MCP calls. This creates a 3-level task hierarchy without generating an intermediate workflow, significantly reducing token consumption.

Yes, Lark MCP integration supports parsing existing Markdown test docs. It extracts test overviews, scenarios, and tasks into a minimal JSON data model to create a 3-level Lark task hierarchy and returns a structured summary with source file references.