Deepseek MCP Server
未申請MCP server for DeepSeek V4 (v4-flash, v4-pro, 1M context): chat, reasoning, function calling, thinking mode, cost tracking. deepseek-chat/reasoner aliases supported.
インストール
claude mcp add -s user deepseek npx @arikusi/deepseek-mcp-server -e DEEPSEEK_API_KEY=your-key-here未申請リスティング
このMCPサーバーはあなたのものですか？
このリスティングは公開情報から自動的にインデックスされました。申請することで、ページの編集、互換性の設定、成長ツールのアンロックができます。2分以内に完了します。
このサーバーを申請するツール（25個）
action
(required): "list" "clear" "delete"
content
Message text
deepseek_chat
Chat with DeepSeek AI models with automatic cost tracking and function calling support.
deepseek_fim
Fill-in-the-Middle completion. You give a prompt (the prefix) and an optional suffix, and the model completes the text in between. It is built for code completion and content infilling rather than conversation. FIM runs on DeepSeek's Beta endpoint in non-thinking mode, and the API caps output at 409
deepseek_sessions
Manage conversation sessions.
json_mode
(optional): Enable JSON output mode (supported by both models)
max_tokens
(optional): Maximum tokens to generate (V4 models support up to 384000)
messages
(required): Array of conversation messages
model
(optional): "deepseek-v4-flash" (default) or "deepseek-v4-pro". The deprecated "deepseek-chat" and "deepseek-reasoner" aliases are still accepted and resolve to v4-flash (non-thinking / thinking); prefer the V4 names.
prompt
(required): The prefix text before the gap. For code completion, this is the code up to the cursor.
reasoning_effort
(optional): "high" (default) or "max", applies only while thinking mode is active
response_schema
(optional): A JSON Schema to validate the model output against. Implies JSON output. The server validates the parsed result and, on failure, issues up to RESPONSESCHEMAMAXRETRIES repair retries (default 2, set 0 to disable) that feed the validation error back to the model. Schema regex patterns are
role
"system" "user" "assistant" "tool"
session_id
(optional): Session ID for multi-turn conversations. Previous context is automatically prepended.
stop
(optional): A stop string or an array of up to 16 stop strings.
stream
(optional): Enable streaming mode (default: false)
structuredContent.effective
and fallback: what was actually sent after alias/thinking resolution, and any silent model fallback that fired
structuredContent.request
a self-contained per-request usage and cost summary (token counts, cache hit/miss, costusd), aggregated across any repair retries
structuredContent.schema
when responseschema is used, {valid, attempts, error?}; jsonparseerror when JSON output could not be recovered
suffix
(optional): The text after the gap. The model fills the space between prompt and suffix.
temperature
(optional): 0-2, controls randomness (default: 1.0). Ignored when thinking mode is enabled.
thinking
(optional): Toggle thinking mode, {type: "enabled"} to reason or {type: "disabled"} for a fast answer (non-thinking is the default)
tool_call_id
(optional): Required for tool role messages
tool_choice
(optional): "auto" "none" "required" {type: "function", function: {name: "..."}}
tools
(optional): Array of tool definitions for function calling (max 128)