Log inSubmit server
MCP Cli logo

MCP Cli

by chrishayuk

CLI for testing and interacting with MCP servers through chat, direct commands, scripted modes, and tool execution.

InstallRepositoryWebsite
Full README Compare alternatives Embed widget Claim this listing

Install MCP Cli

Pick your client, then add it in one click or paste the config.

STDIOPython
~/Library/Application Support/Claude/claude_desktop_config.json
claude_desktop_config.json
{
  "mcpServers": {
    "chrishayuk-mcp-cli": {
      "command": "uvx",
      "args": [
        "mcp-server-sqlite"
      ]
    }
  }
}

Settings → Developer → Edit Config, merge this into mcpServers, then fully quit and reopen Claude Desktop.

We couldn’t confirm it starts correctly

We ran the install command but it didn't respond within our test window — this can mean a slow first-time install rather than a real problem.

uvx mcp-server-sqlite

No response to initialize.

This is an experimental automated check and can have false negatives — missing environment variables, a slow cold install, etc. It doesn’t necessarily mean something’s wrong. Last checked 1mo ago.

Overview

chrishayuk/mcp-cli MCP server is a command-line client for interacting with Model Context Protocol servers and their tools. It supports chat, interactive, command, and direct-command modes, with LLM integration through CHUK Tool Processor and CHUK-LLM. Reach for it when you need to test MCP servers, automate tool calls, inspect health, or manage conversations from a terminal; it defaults to local Ollama with the gpt-oss model.

Use cases

  • Test MCP server tools from a terminal
  • Automate MCP calls in Unix pipelines
  • Inspect server health and tool behavior
  • Run and resume multi-step tool plans
  • Manage local LLM conversations with attachments

Key features

  • Chat, interactive, command, and direct-command modes
  • Concurrent MCP tool execution
  • Session persistence and conversation export
  • Virtual-memory conversation management
  • Execution plans with checkpointing and resume
  • Browser dashboard and MCP Apps support

Security & health

What it can reach and how reliably it runs — from our own checks, not the maintainer’s claims.

Not checked yetLast checked Today

We have not completed a health check for this listing yet.

13/30 checks healthy over the last 78d

Supply-chain signal

No high-severity advisories surfaced by our automated scan.

Critical 0High 0Medium 0Low 0

Scanned 8d ago via OSV.dev · mcp-server-sqlite (PyPI)

Adoption & maintenance

Factual signals from GitHub, npm, and our automated checks — not a rating.

GitHub stars2kStargazers on the source repository.
Last commit2mo agoMost recent push to the default branch.
Install checkInconclusiveDidn't respond in our test window — often a slow first install.
Directory activity4 viewsConfig copies, upvotes, and views on AllMCPs.

How MCP Cli works

What chrishayuk/mcp-cli MCP server does

chrishayuk/mcp-cli MCP server provides a terminal interface for connecting to and exercising Model Context Protocol servers. It can discover and invoke server tools directly, or let an LLM select and use tools during a streamed conversation. The project is intended for testing, interactive operation, and scriptable automation rather than for exposing a business-system API of its own.

Four operating styles are described: chat mode for conversational work, interactive mode for command-driven sessions, command mode for Unix pipelines, and direct commands for individual operations. Responses can include Markdown formatting, syntax highlighting, progress indicators, timing information, and tool execution statistics.

How it works

The CLI acts as the orchestration layer between MCP servers, an LLM provider, and the terminal interface. CHUK Tool Processor handles asynchronous tool execution, while CHUK-LLM supplies the unified model-provider layer. The README also identifies CHUK-Term as the terminal UI component.

In chat mode, the client streams model output and can execute multiple tools concurrently while preserving conversation order. Users can interrupt response generation or tool execution with Ctrl+C. Conversation sessions can be saved, loaded, listed, and automatically saved every 10 turns; conversations can also be exported as Markdown or JSON with metadata.

The experimental virtual-memory feature manages conversation context with configurable budgets and modes. Execution plans can represent multi-step tool-call graphs, run independent steps in parallel, support dry runs, persist checkpoints, and resume after interruption. Plans can be created through the /plan command or, when enabled, generated by the model.

Setup and configuration

Runtime behavior can be adjusted with options including --vm, --vm-budget, --vm-mode, --plan-tools, --log-file, --health-interval, --dashboard, and repeatable --attach arguments. Server-specific tool_timeout and init_timeout settings can override global and default timeout values.

Tools and capabilities

The chrishayuk/mcp-cli MCP server includes:

  • Chat, interactive, command, and direct-command modes.
  • Streaming responses, reasoning visibility, concurrent tool calls, and token usage tracking.
  • Image, audio, text, and code attachments through commands, flags, inline file references, or the dashboard.
  • Health checks through /health, with optional background polling and failure diagnostics.
  • Experimental virtual memory inspection through /memory, /vm, or /mem.
  • Plan creation, inspection, execution, deletion, dry runs, checkpointing, and resume operations.
  • Browser-rendered MCP Apps in sandboxed iframes with WebSocket communication.
  • A real-time dashboard with conversation output, tool activity, plan progress, tool discovery, configuration controls, and file uploads.
  • Secret redaction for bearer tokens, API keys, OAuth tokens, and authorization headers in log output.

Limitations and notes

The virtual-memory functionality is explicitly experimental. The material also identifies later roadmap tiers for traces, memory scopes, skills, scheduling, and multi-agent features, so those should not be assumed to be available.

The default local Ollama setup does not require an API key, but using other model providers may require separate credentials or configuration.

Read the full README View source on GitHub

Reviews

No reviews yet — be the first to share how it worked for you.

Frequently Asked Questions about MCP Cli

Yes.

Yes. Command mode is described as Unix-friendly for scriptable automation and pipelines, and direct commands can run individual operations.

Yes. It supports saved sessions and an experimental virtual-memory feature with configurable budgets, modes, eviction, and page inspection.

Yes. The `--dashboard` flag launches a real-time browser dashboard with conversation, tool activity, plan progress, configuration, and file-upload views.

Related MCP servers

All in Other Tools and Integrations All alternatives
Add a badge to your READMEShow it’s listed — live SVG, Markdown or HTML.

AllMCPs Directory Badge

Full Badge Customizer

Showcase your server listing on GitHub or your project documentation. Embed this dynamic SVG badge to highlight official listing status and live engagement.

Badge Style:
Live Dynamic SVG PreviewMCP Cli AllMCPs Directory Badge
Markdown (GitHub README)
[![AllMCPs](https://allmcps.com/api/badge/chrishayuk-mcp-cli?style=directory)](https://allmcps.com/mcp/chrishayuk-mcp-cli)
HTML Embed
<a href="https://allmcps.com/mcp/chrishayuk-mcp-cli"><img src="https://allmcps.com/api/badge/chrishayuk-mcp-cli?style=directory" alt="MCP Cli on AllMCPs" /></a>