Install Deepseek as Subagent in Claude Code
This server doesn't publish a package to a registry, so follow the install steps in its README. Once it runs, register it with Claude Code:
claude mcp add deepseek-as-subagent -- <command to start the server>Frequently asked questions
What does the Deepseek as Subagent MCP server do?
Run DeepSeek as a sub-agent in Claude Code or Codex CLI, with its own agent loop for reading, editing and running commands in a sandboxed workspace.
Is the Deepseek as Subagent MCP server free?
Yes, most likely. Open source and no paid account or API key is mentioned in the listing. Check the provider for current pricing.
Does Deepseek as Subagent run locally or remotely?
Local. Runs on your machine (stdio). You install it with npx, uvx, Docker or from source.
How do I add Deepseek as Subagent to Claude Code?
Follow the install steps in the project's README at https://github.com/PsChina/deepseek-as-subagent, then register it with "claude mcp add".
More ai agents & coding agents MCP servers
View allAI Usage
MohitBansal321/ai-usage-mcp
Reports real token usage and cost from Claude Code and OpenCode data. Nothing leaves your machine.
DataWorks
aliyun/alibabacloud-dataworks-mcp-server
A Model Context Protocol (MCP) server that provides tools for AI, allowing it to interact with the DataWorks Open API through a standardized interface. This implementation is based on the Aliyun Open API and enables AI agents to perform cloud resources operations seamlessly.
arifOS
ariffazil/arifOS
Constitutional AI Governance with 13 enforced floors (F1-F13) and tri-witness consensus for LLMs.
TRIZ
sv/triz-skill
Deterministic, read-only TRIZ: matrix, principles, effects, routing, type-checked function models.
Small Print
gostanos/smallprint-action
Check an MCP server, skill or plugin before your agent uses it: tool text on record, changes graded.
Dotpals
Rikinshah787/dotpals
Checks what local coding agents such as Claude Code, Codex and Cursor really did: files changed, commands run and test results read from test output, not the agent's word. Read-only.