Provides tools to view git status and recent commits, and automatically commit changes made by Aider.
Supports Google Gemini API as a model provider for Aider, allowing code improvements with Gemini models.
Supports OpenAI API as a model provider for Aider, allowing code improvements with OpenAI models.
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., "@AiderMCP improve code quality in src/main.py"
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.
AiderMCP is a Model Context Protocol (MCP) server that integrates Aider AI pair programming into Claude Desktop and Claude Code. Use Aider's powerful code editing capabilities through natural language conversations with Claude.
⚡ Zero-Install with uvx: Run directly from GitHub - no cloning or setup required!
🚀 One-Command Setup: aider-mcp-server --setup for manual installation
🎯 Natural Language Interface: Use Aider through conversational commands
🔧 5 Powerful Tools: Code improvement, feature addition, bug fixing, refactoring, and status checking
🌍 Cross-Platform: Works on macOS, Windows, and Linux
🔐 Multi-Provider: Supports Anthropic, OpenAI, and Google Gemini API keys
Related MCP server: Aider MCP Server
Python 3.11 or 3.12 (Download)
⚠️ Python 3.13 not yet supported
Claude Desktop (Download)
API Key from at least one provider:
Google AI Studio (default)
Git installed and initialized in your project
The fastest way to use AiderMCP - no cloning or installation required!
Edit your Claude Desktop configuration file:
macOS: ~/Library/Application Support/Claude/claude_desktop_config.json
Windows: %APPDATA%\Claude\claude_desktop_config.json
Linux: ~/.config/Claude/claude_desktop_config.json
Add this configuration:
{
"mcpServers": {
"aider": {
"command": "uvx",
"args": [
"--from",
"git+https://github.com/fuzemobi/AiderMCP.git",
"aider-mcp-server"
],
"env": {
"GEMINI_API_KEY": "your-gemini-key-here",
"ANTHROPIC_API_KEY": "your-anthropic-key-here",
"OPENAI_API_KEY": "your-openai-key-here"
}
}
}
}Notes:
Add your API keys (at least one: Gemini, Anthropic, or OpenAI)
If you have other MCP servers, merge this into the existing mcpServers object
If you forked this repo, update the URL to your fork's location
That's it! uvx will automatically:
Download the repository
Install all dependencies (aider-chat, fastmcp)
Start the MCP server
If you prefer manual installation or need to modify the code:
# Clone the repository
git clone https://github.com/fuzemobi/AiderMCP.git
cd AiderMCP
# Create virtual environment (Python 3.11 or 3.12)
python3.12 -m venv .venv
source .venv/bin/activate # Windows: .venv\Scripts\activate
# Install
pip install -e .# macOS/Linux
export GEMINI_API_KEY="your-key-here"
export ANTHROPIC_API_KEY="your-key-here" # optional
export OPENAI_API_KEY="your-key-here" # optional
# Windows PowerShell
$env:GEMINI_API_KEY="your-key-here"aider-mcp-server --setupImportant: Restart Claude Desktop for changes to take effect.
Use natural language to work with Aider:
Improve Code Quality:
Use aider to improve code quality in src/main.pyAdd Features:
Ask aider to add user authentication with JWT tokensFix Bugs:
Have aider fix the connection timeout bug in server.pyRefactor Code:
Get aider to refactor the database module for better performanceCheck Status:
Use aider to show git status and recent commitsTool | Description |
improve_code | Improve code quality, add type hints, enhance error handling |
add_feature | Add new features to your application |
fix_bug | Fix bugs and issues in your codebase |
refactor | Refactor code for better structure and performance |
get_status | View git status and recent commits |
All tools support these optional parameters:
files: Specific files to work on
model: AI model to use (default: gemini/gemini-1.5-pro-latest)
auto_commit: Automatically commit changes (default: True)
project_path: Project directory path (default: current directory)
Aider is an AI pair programming tool that works with your code. It can:
Edit multiple files at once
Make coordinated changes across your codebase
Understand your project structure
Follow coding conventions and patterns
Create git commits automatically
AiderMCP makes Aider accessible through Claude's conversational interface, letting you describe what you want in natural language.
Tools not appearing in Claude Desktop?
Restart Claude Desktop after running setup
Re-run: aider-mcp-server --setup
Check config file exists:
macOS: ~/Library/Application Support/Claude/claude_desktop_config.json
Windows: %APPDATA%\Claude\claude_desktop_config.json
Linux: ~/.config/Claude/claude_desktop_config.json
Python version errors?
python3 --version # Must be 3.11 or 3.12
# Recreate venv if needed:
rm -rf .venv
python3.12 -m venv .venv
source .venv/bin/activate
pip install -e .API key issues?
# Verify keys are set
echo $GEMINI_API_KEY
# Or add to config file manually after running setupMIT License - see LICENSE file for details.
Made with ❤️ for developers who want AI-powered coding in Claude
Add a new feature using Aider
| Name | Required | Description | Default |
|---|---|---|---|
| files | No | Files to modify or create | |
| model | No | AI model to use (default: gemini/gemini-1.5-pro-latest) | gemini/gemini-1.5-pro-latest |
| project_path | No | Absolute path to project directory (default: current directory) | . |
| feature_description | Yes | Description of the feature to add |
| Name | Required | Description |
|---|---|---|
| result | Yes |
Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?
No annotations are provided, so the description must carry the behavioral disclosure burden. It only says 'Add a new feature using Aider,' which implies mutation but gives no detail on side effects, permissions, or failure modes.
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?
A single front-loaded sentence with no wasted words. It is efficiently concise, even if it sacrifices depth for brevity.
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?
Despite full schema coverage and an output schema, the description omits crucial context: when to use add_feature versus improve_code or fix_bug, and what using Aider entails behaviorally. The tool is complex enough that this minimal description leaves major gaps.
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%, with all four parameters having descriptions. The tool description adds no parameter semantics, so the baseline score of 3 is appropriate.
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 verb 'Add' and resource 'feature' clearly state the action. Adding 'using Aider' provides method context, distinguishing it from siblings like improve_code, fix_bug, and refactor.
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 is given on when to use this tool over its siblings. There is no mention of prerequisites, alternatives, or contexts where other tools would be more appropriate.
Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.
Fix a bug using Aider
| Name | Required | Description | Default |
|---|---|---|---|
| files | No | Files that may contain the bug | |
| project_path | No | Absolute path to project directory (default: current directory) | . |
| error_message | No | Error message if available | |
| bug_description | Yes | Description of the bug to fix |
| Name | Required | Description |
|---|---|---|
| result | Yes |
Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?
With no annotations provided, the description must disclose behavioral traits. It only mentions 'using Aider,' which implies an external dependency but doesn't state that the tool modifies code, may require git access, or could have side effects on the file system. There is no mention of permissions, reversibility, or failure modes. This is insufficient for a mutation 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?
The description is extremely concise at six words with no wasted text. However, it is so brief that it lacks substance, which hurts other dimensions. For conciseness alone, it earns credit for being efficient, but it is not structured or informative enough to be a model of good tool descriptions.
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 the tool's complexity (4 parameters, mutation behavior) and lack of annotations, the description is far from complete. It does not explain what Aider is, how the bug will be fixed, which files are affected, or what the output will be. The presence of an output schema reduces the need to describe return values, but the description still leaves major gaps about usage and effects.
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 all parameters have descriptions. The tool description itself adds no additional parameter semantics beyond what the schema already provides. Baseline of 3 is appropriate since the schema does the heavy lifting and no extra context is given.
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 verb 'fix' combined with resource 'bug' clearly indicates the tool's purpose. It distinguishes from siblings like add_feature, refactor, and improve_code by targeting bugs specifically. However, 'using Aider' may be unclear if the agent doesn't know Aider, and it lacks additional context about the scope or manner of fixing.
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 does not provide any guidance on when to use this tool versus alternatives. It merely states what it does, without specifying conditions (e.g., 'use when there is a bug') or exclusions (e.g., 'for new functionality use add_feature'). No naming of sibling tools or decision criteria.
Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.
Get Aider and git status
| Name | Required | Description | Default |
|---|---|---|---|
| project_path | No | Absolute path to project directory (default: current directory) | . |
| Name | Required | Description |
|---|---|---|
| result | Yes |
Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?
There are no annotations, so the description carries the full burden. The word 'Get' implies a read-only operation and no side effects, but this is not explicitly stated, and the tool's behavior (e.g., whether it checks for changes, current branch, or commits) is undisclosed. The output schema exists but the description does not reference it.
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 concise sentence that captures the tool's essence without wordiness. Every word contributes to the meaning.
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 simple status-checking tool with one parameter and an output schema, the description is minimal but arguably sufficient for a basic understanding. However, it lacks context about what 'Aider and git status' specifically reports, and the absence of usage guidance makes it less complete for an agent deciding when to invoke it.
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 single parameter project_path is fully documented in the schema with a clear description and default value. The description does not add extra detail, but with 100% schema coverage, the baseline of 3 is appropriate.
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 uses the verb 'Get' with the resource 'Aider and git status', making the action unambiguous. It is distinct from sibling tools which are code modification actions (improve_code, add_feature, etc.), though it does not explicitly describe what 'Aider and git status' entails beyond the name.
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 is provided on when to use this tool versus alternatives. The sibling tools are all action-oriented, so one might infer this is for pre-action status checks, but the description does not state this or offer any alternatives.
Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.
Use Aider to improve code quality, fix bugs, or implement features
| Name | Required | Description | Default |
|---|---|---|---|
| files | No | List of files to work on | |
| model | No | AI model to use (default: gemini/gemini-1.5-pro-latest) | gemini/gemini-1.5-pro-latest |
| auto_commit | No | Automatically commit changes (default: True) | |
| instruction | Yes | What to improve or implement | |
| project_path | No | Absolute path to project directory (default: current directory) | . |
| Name | Required | Description |
|---|---|---|
| result | Yes |
Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?
There are no annotations provided, so the description must carry the full burden of behavioral disclosure. It fails to mention that this tool modifies code files, may commit changes (auto_commit param), or has any side effects, leaving the agent without critical safety information.
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, front-loading the core purpose. It is concise and well-structured, though this brevity contributes to gaps in other dimensions.
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?
Despite the rich schema and output schema, the description omits essential context for a code-modifying tool. There is no mention of side effects, commit behavior, or filesystem interaction, and with no annotations, the description is insufficient for safe use.
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 the schema already documents all five parameters (files, model, auto_commit, instruction, project_path). The description adds no parameter-specific meaning beyond the general purpose, so the baseline of 3 is appropriate.
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's action ('improve code quality, fix bugs, or implement features') using Aider, which gives a specific verb and resource. However, it overlaps with sibling tools like fix_bug, add_feature, and refactor without distinguishing itself, so it doesn't fully differentiate.
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 is provided on when to use this tool versus alternatives. The description even includes 'fix bugs' and 'implement features', which are covered by sibling tools, but it does not explain when to choose this tool or when to choose a sibling.
Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.
Refactor code using Aider
| Name | Required | Description | Default |
|---|---|---|---|
| files | No | Files to refactor | |
| project_path | No | Absolute path to project directory (default: current directory) | . |
| refactor_goal | Yes | What to refactor and why |
| Name | Required | Description |
|---|---|---|
| result | Yes |
Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?
With no annotations, the description must carry the burden of behavioral disclosure. It only mentions 'using Aider', hinting at an external tool, but does not explain side effects (e.g., modifying files), required permissions, or how the refactoring is executed.
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, front-loaded sentence with no wasted words. However, it is under-specified for a tool of this complexity, so while concise, it is not fully appropriate in length.
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?
Despite having an output schema and sufficient parameter documentation, the description lacks essential context: when to use it, what behavior to expect, and how it differs from sibling tools. The agent is left without enough information to select or invoke this tool confidently.
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 the parameters are already well documented in the input schema. The description adds no additional semantic meaning beyond what the schema provides, which meets the baseline for high coverage.
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 uses a specific verb ('refactor') and names the resource ('code'), making the core action clear. However, it does not differentiate from sibling tools like improve_code, and 'using Aider' adds an implementation detail rather than clarifying scope.
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 is provided about when to use this tool versus alternatives such as improve_code, add_feature, or fix_bug. There is no mention of prerequisites, exclusions, or context in which refactoring is appropriate.
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.
v2.0.0add_featurefix_bugget_statusimprove_coderefactorimprove_code overlaps significantly with add_feature and fix_bug, as it explicitly includes both fixing bugs and implementing features. This makes it unclear whether an agent should use the generic tool or the specific one. refactor is somewhat distinct but still a code-modification action, adding to the ambiguity.
Most tool names follow a consistent verb_noun pattern (improve_code, add_feature, fix_bug, get_status). The exception is 'refactor', a bare verb, which is a minor deviation but does not significantly harm predictability.
With 5 tools, the server is well-scoped for its purpose of interfacing with Aider. Each tool maps to a distinct type of operation, and the count is neither too thin nor bloated.
The tool set covers the main code-modification workflows (improve, add feature, fix bug, refactor) plus a status check. However, the generic improve_code makes the specific tools redundant, and there are no tools for reverting or committing changes, leaving minor gaps.
Unclaimed servers have limited discoverability.
If you are the server author, to access and configure the admin panel.
Use AI models for chat, image, and video generation from Claude Code and other MCP hosts.
Source-checked CLI guides and model-aware planning for Claude Code, Codex, and Grok Build.
Claude-powered AI tools: research, write, code, analyze, translate, debate, pitch, score, and more.
Persistent context for Claude. Your AI always knows your projects and next actions across sessions.
We provide all the information about MCP servers via our MCP API.
curl -X GET 'https://glama.ai/api/mcp/v1/servers/fuzemobi/AiderMCP'
If you have feedback or need assistance with the MCP directory API, please join our Discord server
