Offload cheap work from your AI agent to a local Ollama model, at zero cloud cost.
Offload cheap work from your AI agent to a local Ollama model, at zero cloud cost.
This isn't a generic "wrap the Ollama API" server. Each tool ships with a baked-in system prompt and a description written for the calling agent, so the agent knows when to hand off and gets a tuned result back without re-stating instructions every call.
ollama serve) with at least one model pulled, e.g. ollama pull qwen2.5-coder:14b. - Python 3.11+ (or just uvx, which manages it for you).Compare MCP servers by category, or browse hosted MCP servers if you prefer an endpoint over a local process.
Check the linked publisher documentation and your connected client's tool list. Names and availability can vary with the server version and account permissions.
| Tool | What it does |
|---|---|
| ask_local | A single prompt with an optional system instruction |
| chat_local | A conversation with explicit message history |
| summarize_local | Summaries of supplied text, optionally focused on a topic |
| code_review_local | Initial review of supplied code or a diff |
| draft_commit_message_local | A commit message from a supplied diff |
| extract_local | Extracting items such as URLs, names, or error codes |
| list_models | Discovering installed Ollama models |
| server_info | Inspecting effective server configuration |
{
"mcpServers": {
"ollama-handoff": {
"command": "uvx",
"args": [
"ollama-handoff@0.1.3"
],
"env": {
"OLLAMA_DEFAULT_MODEL": "llama3.1:8b"
}
}
}
}From the project documentation. Replace placeholders and example paths before connecting.
ollama serve) with at least one model pulled, e.g. ollama pull qwen2.5-coder:14b. - Python 3.11+ (or just uvx, which manages it for you).| Variable | Description | Required |
|---|---|---|
| OLLAMA_DEFAULT_MODEL | Declared in the maintainer’s client configuration. | Check documentation |
Read, search and edit files inside allowed directories with the Model Context Protocol reference Filesystem server.
Retrieve version-aware library documentation and code examples with Context7 MCP, using resolve-library-id and query-docs.
Feishu/Lark OpenAPI MCP
On first use, macOS prompts for **Calendar** and **Reminders** access — click **Allow**. Building from source, upgrading in place, or running under
Bounded Scala compiler, build, test, type, effect, symbol, and SemanticDB evidence for agents.
askme-cli MCP server that collects user's next plan or confirmation through terminal window