Let a planning agent hand bounded language work to a local Ollama model.
Connect Runs on your machine
claude mcp add llm -e SEZZLEE_LLM_MODEL=<SEZZLEE_LLM_MODEL> -e SEZZLEE_LLM_BASE_URL=<SEZZLEE_LLM_BASE_URL> -e SEZZLEE_LLM_NUM_CTX=<SEZZLEE_LLM_NUM_CTX> -e SEZZLEE_LLM_TIMEOUT_MS=<SEZZLEE_LLM_TIMEOUT_MS> -e SEZZLEE_LLM_KEEP_ALIVE=<SEZZLEE_LLM_KEEP_ALIVE> -e SEZZLEE_LLM_ROOT=<SEZZLEE_LLM_ROOT> -e SEZZLEE_LLM_OUTPUT_DIR=<SEZZLEE_LLM_OUTPUT_DIR> -- npx -y @sezzlee/[email protected]{
"mcpServers": {
"llm": {
"command": "npx",
"args": [
"-y",
"@sezzlee/[email protected]"
],
"env": {
"SEZZLEE_LLM_MODEL": "<SEZZLEE_LLM_MODEL>",
"SEZZLEE_LLM_BASE_URL": "<SEZZLEE_LLM_BASE_URL>",
"SEZZLEE_LLM_NUM_CTX": "<SEZZLEE_LLM_NUM_CTX>",
"SEZZLEE_LLM_TIMEOUT_MS": "<SEZZLEE_LLM_TIMEOUT_MS>",
"SEZZLEE_LLM_KEEP_ALIVE": "<SEZZLEE_LLM_KEEP_ALIVE>",
"SEZZLEE_LLM_ROOT": "<SEZZLEE_LLM_ROOT>",
"SEZZLEE_LLM_OUTPUT_DIR": "<SEZZLEE_LLM_OUTPUT_DIR>"
}
}
}
}{
"mcpServers": {
"llm": {
"command": "npx",
"args": [
"-y",
"@sezzlee/[email protected]"
],
"env": {
"SEZZLEE_LLM_MODEL": "<SEZZLEE_LLM_MODEL>",
"SEZZLEE_LLM_BASE_URL": "<SEZZLEE_LLM_BASE_URL>",
"SEZZLEE_LLM_NUM_CTX": "<SEZZLEE_LLM_NUM_CTX>",
"SEZZLEE_LLM_TIMEOUT_MS": "<SEZZLEE_LLM_TIMEOUT_MS>",
"SEZZLEE_LLM_KEEP_ALIVE": "<SEZZLEE_LLM_KEEP_ALIVE>",
"SEZZLEE_LLM_ROOT": "<SEZZLEE_LLM_ROOT>",
"SEZZLEE_LLM_OUTPUT_DIR": "<SEZZLEE_LLM_OUTPUT_DIR>"
}
}
}
}{
"servers": {
"llm": {
"type": "stdio",
"command": "npx",
"args": [
"-y",
"@sezzlee/[email protected]"
],
"env": {
"SEZZLEE_LLM_MODEL": "<SEZZLEE_LLM_MODEL>",
"SEZZLEE_LLM_BASE_URL": "<SEZZLEE_LLM_BASE_URL>",
"SEZZLEE_LLM_NUM_CTX": "<SEZZLEE_LLM_NUM_CTX>",
"SEZZLEE_LLM_TIMEOUT_MS": "<SEZZLEE_LLM_TIMEOUT_MS>",
"SEZZLEE_LLM_KEEP_ALIVE": "<SEZZLEE_LLM_KEEP_ALIVE>",
"SEZZLEE_LLM_ROOT": "<SEZZLEE_LLM_ROOT>",
"SEZZLEE_LLM_OUTPUT_DIR": "<SEZZLEE_LLM_OUTPUT_DIR>"
}
}
}
}Replace the values in angle brackets with your own. Keys stay on your machine; we never see them.
Tools
This server runs locally, so its tools are listed by your MCP client once it starts.
Environment variables
| Name | Description |
|---|---|
SEZZLEE_LLM_MODELRequired | Ollama model to call |
SEZZLEE_LLM_BASE_URLOptional | Ollama base URL |
SEZZLEE_LLM_NUM_CTXOptional | Context window in tokens |
SEZZLEE_LLM_TIMEOUT_MSOptional | Per-call timeout in milliseconds |
SEZZLEE_LLM_KEEP_ALIVEOptional | How long Ollama keeps the model loaded |
SEZZLEE_LLM_ROOTOptional | Workspace folder paths resolve against |
SEZZLEE_LLM_OUTPUT_DIROptional | Folder local_map writes its CSV into |
Details
| Package | @sezzlee/[email protected] (NPM, stdio) |
| First published | 11 Oct 2026 |
| Last updated | 11 Oct 2026 |
| Registry name | io.github.sezzlee/llm |
More ai and media servers See all
Listing data from the official MCP registry. Status and tools from NestDaddy's daily check. Browse all MCP servers.