Sezzlee Local LLM MCP Server
Let a planning agent hand bounded language work to a local Ollama model.
Install
{
"mcpServers": {
"sezzlee-llm-mcp": {
"command": "npx",
"args": [
"-y",
"@sezzlee/llm-mcp@0.1.3"
],
"env": {
"SEZZLEE_LLM_MODEL": "",
"SEZZLEE_LLM_BASE_URL": "http://127.0.0.1:11434",
"SEZZLEE_LLM_NUM_CTX": "16384",
"SEZZLEE_LLM_TIMEOUT_MS": "300000",
"SEZZLEE_LLM_KEEP_ALIVE": "",
"SEZZLEE_LLM_ROOT": "",
"SEZZLEE_LLM_OUTPUT_DIR": ""
}
}
}
}Related servers
Codebase Memory
77GoodCodebase knowledge graph for AI agents — 162 languages, sub-ms queries, 99% fewer tokens.
haiku.rag
74GoodLocal-first agentic RAG with citations - hybrid search, reranking, multimodal document retrieval