deepseek-delegate-mcp
Server Configuration
Describes the environment variables required to run the server.
| Name | Required | Description | Default |
|---|---|---|---|
| DEEPSEEK_MODEL | No | Default model for the tool. Use 'deepseek-v4-flash' or 'deepseek-v4-pro'. | deepseek-v4-flash |
| DEEPSEEK_API_KEY | No | DeepSeek API key. Falls back to your Codex config. Looked up in order: DEEPSEEK_API_KEY environment variable, then experimental_bearer_token under [model_providers.deepseek] in ~/.codex/config.toml. | |
| DEEPSEEK_BASE_URL | No | API base URL. | https://api.deepseek.com |
| DEEPSEEK_THINKING | No | Thinking mode default: 'enabled' or 'disabled'. Disabled is fast/cheap. | disabled |
| DEEPSEEK_WIRE_API | No | API style: 'responses' (native Responses API) or 'chat' (OpenAI-style chat completions). | responses |
| DEEPSEEK_MAX_PROMPT_CHARS | No | Maximum prompt length in characters. Larger prompts are rejected to prevent giant-payload delegation. | 30000 |
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 | {
"listChanged": true
} |
Tools
Functions exposed to the LLM to take actions
| Name | Description |
|---|---|
| call_deepseek_sub_agentA | Delegates massive text processing, code generation boilerplate, or complex sub-modules to DeepSeek V4 to conserve Sol context tokens. Returns the model's text output. |
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 1 tool
With only one tool, there is no possibility of confusing it with others. The tool's purpose is clearly defined as delegating to DeepSeek V4.
The tool name follows a clear verb_noun pattern ('call' + 'deepseek_sub_agent'). With only one tool, there are no inconsistencies or mixed conventions.
The server has a single tool, which is slightly below the typical well-scoped range. However, it is appropriate for the narrow purpose of delegating to DeepSeek, making the count reasonable.
The tool provides a generic delegation capability for any text processing or code generation task, covering the entire stated domain. There are no obvious missing operations for a simple delegate service.