Model Worker MCP
Server Configuration
Describes the environment variables required to run the server.
| Name | Required | Description | Default |
|---|---|---|---|
| MODEL_WORKER_MODEL | No | The model to use. Default primary model is `qwen3.7-plus`; single-file small changes may switch to `qwen3.6-flash`. `deepseek-flash-0731` is used automatically for medium/low-risk logic, bug fixes, tests, and small backend features, and can also be specified manually. | qwen3.7-plus |
| MODEL_WORKER_API_KEY | Yes | API key for the OpenAI-compatible model provider. Set it in the user environment variables, or store it encrypted (Windows DPAPI) by running `npm run credentials:set -- worker-primary`. The key never appears in config files, task parameters, diffs, or logs. | |
| MODEL_WORKER_BASE_URL | No | Any OpenAI-compatible `/chat/completions` endpoint. Defaults to Alibaba Cloud Bailian (https://dashscope.aliyuncs.com/compatible-mode/v1); if you use a Bailian workspace-specific endpoint, set your own workspace URL. | https://dashscope.aliyuncs.com/compatible-mode/v1 |
Instructions
Guidance the server publishes about itself, which clients place ahead of the tool catalog so the model reads it before choosing anything.
This server publishes no instructions, or was last inspected before Glama recorded them.
Capabilities
Features and capabilities supported by this server
Protocol revision2025-11-25
| Capability | Details |
|---|---|
| tools | {} |
Tools
Functions exposed to the LLM to take actions
| Name | Description |
|---|---|
| route_code_taskC | Classify a coding task by risk and automatically delegate safe work to the discounted model tier. |
| delegate_code_taskC | Run a bounded coding task through the configured provider. |
| get_task_statusB | Get status, diff and validation output for a delegated task. |
| cancel_taskC | Cancel a running delegated task. |
| get_usage_summaryA | Get persisted token usage and estimated CNY cost, optionally filtered by model or time range. |
Prompts
Interactive templates invoked by user choice
| Name | Description |
|---|---|
No prompts | |
Resources
Contextual data attached and managed by the client
| Name | Description |
|---|---|
No resources | |
TDQS
Scored across 5 tools
route_code_task and delegate_code_task both initiate delegated coding work, which could cause some initial confusion, but the descriptions clarify that one performs risk classification and auto-routing while the other directly runs a bounded task. The remaining tools (status, cancel, usage) are clearly distinct.
Most tools follow a verb_noun pattern (route_code_task, delegate_code_task, get_task_status, get_usage_summary), but cancel_task breaks the pattern by omitting the 'code' qualifier used by the delegation tools. The set is still readable and predictable overall.
Five tools is well-scoped for a task delegation worker: routing, execution, status, cancellation, and usage reporting. Each tool clearly earns its place with no redundancy.
The core lifecycle is covered: initiate/route, run, check status, cancel, and inspect usage. Minor gaps exist, such as no list/history operation to enumerate past tasks or a retry/resume capability, but these are workable around.