Click on "Install Server".
Wait a few minutes for the server to deploy. Once ready, it will show a "Started" state.
In the chat, type @ followed by the MCP server name and your instructions, e.g., "@tabby-mcp list all open tabs in Tabby"
That's it! The server will respond to your query, and you can continue using it as needed.
Here is a step-by-step guide with screenshots.
A lightweight MCP (Model Context Protocol) server that enables control of the Tabby terminal via Chrome DevTools Protocol (CDP).
list_targets - List CDP targets (tabs) with index, title, url, ws_url
execute_js - Execute JavaScript code in Tabby's Electron context
query - Query DOM elements by CSS selector (auto-waits for Angular)
screenshot - Capture Tabby window screenshot
Related MCP server: MCP Browser Logger
This MCP server was built to automate development and testing of TabbySpaces.
TabbySpaces was written 100% by Claude Code - Igor didn't look at the Angular code once. But for this Python project, he has MCP experience and actually reviewed the code.
Here's the fun part: Claude Code (running inside TabbySpaces) served as the tester for this project. The AI that built the plugin now uses this MCP server to test it. Full circle.
Python 3.10+
uv package manager
Tabby terminal with CDP debugging enabled
pip install tabby-mcpOr with uv:
uv pip install tabby-mcpgit clone https://github.com/halilc4/tabby-mcp.git
cd tabby-mcp
uv synctabby.exe --remote-debugging-port=9222Add to your Claude Code MCP settings (~/.claude/claude_desktop_config.json):
{
"mcpServers": {
"tabby": {
"command": "uv",
"args": ["--directory", "/path/to/tabby-mcp", "run", "tabby-mcp"]
}
}
}Once configured, Claude Code can interact with Tabby through MCP tools:
# Execute JavaScript
execute_js({ "code": "document.title" })
# Query DOM elements
query({ "selector": ".tab-bar button" })server.py -> tools.py -> cdp.py -> Tabby (CDP port 9222)Module | Purpose |
| MCP server entry point with stdio transport |
| MCP tool definitions and handlers |
| CDP connection management via pychrome |
MIT
Execute JavaScript code in Tabby terminal context and return the result. Code is wrapped in async IIFE by default for fresh scope and await support.
| Name | Required | Description | Default |
|---|---|---|---|
| target | Yes | Target tab: index (0=first, -1=last) or WebSocket URL from list_targets | |
| code | Yes | JavaScript code to execute. Use 'return' to return values. | |
| wrap | No | Wrap code in async IIFE for fresh scope + await support. Set to false for raw execution (e.g., defining globals). |
Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?
Without annotations, the description must carry behavioral disclosure. It correctly reveals default async IIFE wrapping and await support. However, it omits details on error handling, side effects, or security implications, which are relevant for a code execution tool.
Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.
Is the description appropriately sized, front-loaded, and free of redundancy?
Two efficient sentences: the first front-loads the primary purpose, the second adds critical behavioral detail. No filler, every word contributes.
Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.
Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?
For a 3-parameter tool with no output schema or annotations, the description is moderately complete. It covers the main action and the wrap behavior, but lacks information on error handling, return format details, and potential side effects, which are needed given the tool's power.
Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.
Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?
Schema coverage is 100%, so baseline is 3. The description adds context by explaining the wrap parameter's purpose (async IIFE for fresh scope and await support), going beyond the schema. However, it does not add meaning to target or code beyond what the schema provides.
Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.
Does the description clearly state what the tool does and how it differs from similar tools?
The description clearly states that the tool executes JavaScript code in the Tabby terminal context and returns results, distinguishing it from siblings like list_targets (listing) and screenshot (capturing). The mention of async IIFE adds specificity.
Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.
Does the description explain when to use this tool, when not to, or what alternatives exist?
The description implies when to use this tool (for executing JS code) versus siblings, but lacks explicit when-not or alternative guidance. The context from sibling names helps, but the description itself doesn't provide exclusion criteria.
Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.
List available CDP targets (tabs) with their index, URL, and WebSocket URL
| Name | Required | Description | Default |
|---|---|---|---|
No parameters | |||
Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?
No annotations are provided, so the description carries the full burden. It accurately describes a read-only operation returning target details, but does not disclose any side effects, permission requirements, or behavior in edge cases (e.g., no targets). The description is truthful but minimally transparent.
Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.
Is the description appropriately sized, front-loaded, and free of redundancy?
The description is a single sentence with no wasted words. It front-loads the action and immediately conveys what the tool does and what it returns. Achieves high conciseness while remaining clear.
Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.
Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?
Given zero parameters and no output schema, the description adequately specifies the tool's purpose and the information returned (index, URL, WebSocket URL). It could be improved by noting the return format (e.g., array of objects), but still sufficient for a simple list tool.
Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.
Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?
The tool has zero parameters and the schema coverage is 100% (vacuously). The description does not need to add param information. Baseline score of 4 is appropriate since nothing is missing.
Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.
Does the description clearly state what the tool does and how it differs from similar tools?
The description clearly states the tool lists available CDP targets (tabs) and specifies the returned information (index, URL, WebSocket URL). It uses a specific verb+resource structure and distinguishes well from sibling tools like execute_js, query, and screenshot, which serve different purposes.
Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.
Does the description explain when to use this tool, when not to, or what alternatives exist?
The description provides no explicit guidance on when to use this tool versus alternatives or any prerequisites. It only states what the tool does, leaving the agent to infer usage context. For a simple enumeration tool with no parameters, this is a noticeable gap.
Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.
Query DOM elements by CSS selector. Automatically waits for Angular and element to exist.
| Name | Required | Description | Default |
|---|---|---|---|
| target | Yes | Target tab: index (0=first, -1=last) or WebSocket URL from list_targets | |
| selector | Yes | CSS selector to query | |
| include_children | No | Include children preview (first 10, with tagName, id, className) | |
| include_text | No | Include textContent (truncated to 200 chars) | |
| skip_wait | No | Skip Angular/element wait (use when element definitely exists) |
Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?
No annotations are provided, so the description carries the full burden. It discloses automatic waiting behavior, but does not mention error handling, return value format, or what happens if the element is not found. This leaves notable gaps in transparency.
Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.
Is the description appropriately sized, front-loaded, and free of redundancy?
The description is two sentences long, front-loading the purpose and adding key behavioral info without any fluff. Every sentence earns its place.
Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.
Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?
Given no output schema, the description should explain the return value and structure, but it does not. It also lacks error handling details and guidance among siblings, making it incomplete for a tool with 5 parameters and no annotations.
Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.
Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?
Schema coverage is 100%, so baseline is 3. The description adds no extra meaning beyond the parameter descriptions already in the schema, providing no additional semantic value.
Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.
Does the description clearly state what the tool does and how it differs from similar tools?
The description clearly states 'Query DOM elements by CSS selector,' specifying the verb (query), resource (DOM elements), and mechanism (CSS selector). This distinguishes it from sibling tools like execute_js (execute arbitrary JS), list_targets (list available targets), and screenshot (capture image).
Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.
Does the description explain when to use this tool, when not to, or what alternatives exist?
The description says 'Automatically waits for Angular and element to exist,' which implies it is ideal when elements may load asynchronously. However, it does not explicitly state when not to use it or compare it to siblings, missing explicit exclusion guidance.
Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.
Capture screenshot of Tabby terminal window or specific element
| Name | Required | Description | Default |
|---|---|---|---|
| target | Yes | Target tab: index (0=first, -1=last) or WebSocket URL from list_targets | |
| selector | No | CSS selector for element screenshot (optional, full window if omitted) | |
| format | No | Image format | jpeg |
| quality | No | JPEG quality (ignored for PNG) | |
| save_path | No | File path to save screenshot (optional, returns base64 if omitted) |
Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?
No annotations provided; description does not disclose any behavioral traits (e.g., permissions, side effects). It only states the basic function.
Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.
Is the description appropriately sized, front-loaded, and free of redundancy?
Single sentence with no wasted words, but could benefit from more structure or context without being verbose.
Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.
Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?
Given 5 parameters, no output schema, and no annotations, the description is too minimal—fails to explain return behavior or parameter usage beyond what schema provides.
Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.
Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?
Schema description coverage is 100%, so baseline 3. Description adds minimal meaning beyond schema (e.g., 'window or element' maps to target and selector).
Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.
Does the description clearly state what the tool does and how it differs from similar tools?
Description clearly states 'Capture screenshot of Tabby terminal window or specific element', which is a specific verb-resource pair that distinguishes it from sibling tools (execute_js, list_targets, query).
Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.
Does the description explain when to use this tool, when not to, or what alternatives exist?
No guidance on when to use this tool versus alternatives, no mention of prerequisites or exclusions.
Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.
Recent tool additions, removals, and schema changes observed during successful MCP inspections. Dates show when Glama detected each change.
v0.1.1execute_jslist_targetsqueryscreenshotEach tool has a clearly distinct purpose: execute_js runs code, list_targets lists tabs, query finds DOM elements, screenshot captures images. There is no overlap.
Most tools follow a verb_noun pattern (execute_js, list_targets), but 'query' and 'screenshot' are less consistent. Still, the pattern is understandable.
Four tools is appropriate for a focused automation toolkit. Each tool serves a clear function without being excessive or insufficient.
The set covers basic inspection and execution but lacks common browser actions like navigation, clicking, or typing, which limits its completeness for typical automation tasks.
Unclaimed servers have limited discoverability.
If you are the server author, to access and configure the admin panel.
MCP server to assist with JxBrowser development.
MCP server for understanding Javascript internals from ECMAScript specification.
A TypeScript MCP server for Home Assistant, enabling programmatic management of entities, automati…
We provide all the information about MCP servers via our MCP API.
curl -X GET 'https://glama.ai/api/mcp/v1/servers/halilc4/tabby-mcp'
If you have feedback or need assistance with the MCP directory API, please join our Discord server
