Mintplex-Labs/anything-llm
1
0
Fork 0
mirror of https://github.com/Mintplex-Labs/anything-llm.git synced 2026-07-19 14:13:35 -04:00
Code Issues 328 Packages Projects Releases 10 Wiki Activity

[GH-ISSUE #2883] [FEAT]: Model Context Protocol (MCP) Integration #1834

New Issue
Closed
opened 2026-02-22 18:26:44 -05:00 by yindo · 7 comments
yindo commented 2026-02-22 18:26:44 -05:00
Owner
Copy Link
Copy Source

Originally created by @GitDakky on GitHub (Dec 20, 2024).
Original GitHub issue: https://github.com/Mintplex-Labs/anything-llm/issues/2883

What would you like to see?

Description

Request to integrate Model Context Protocol (MCP) support into AnythingLLM to enhance interoperability and standardization of context handling across different LLM implementations.

Motivation

MCP is becoming a standard protocol for AI applications, offering a "USB-C port for AI applications" that would benefit AnythingLLM by:

  • Providing standardized context handling across different LLM providers
  • Enabling easier integration with other MCP-compatible tools and services
  • Improving the flexibility of switching between different LLM providers
  • Adding support for standardized prompt templates and tool interactions

Proposed Implementation

The implementation would require:

1. Core MCP Features

  • Resource handling for data access and context management
  • Prompt template support with argument handling
  • Tool integration capabilities
  • Support for both synchronous and asynchronous operations

2. Integration Points

  • MCP client implementation for connecting to MCP servers
  • Server configuration support
  • Protocol handlers for MCP communication
  • Resource update subscription mechanisms

3. Security Considerations

  • Proper authentication and authorization
  • Secure handling of API keys and sensitive information
  • Data security best practices

Benefits

  1. Standardization: Align with emerging industry standards for LLM context handling
  2. Interoperability: Better integration with other AI tools and services
  3. Flexibility: Easier switching between different LLM providers
  4. Extensibility: Simplified addition of new features and capabilities
  5. Community: Access to growing ecosystem of MCP-compatible tools

Technical Requirements

  • Implementation of MCP client capabilities
  • Support for MCP protocol specifications
  • Integration with existing AnythingLLM architecture
  • Backward compatibility with current features

Additional Context

MCP is being adopted by various applications including:

  • Claude Desktop
  • Sourcegraph Cody
  • Continue
  • Zed
  • Firebase Genkit
Originally created by @GitDakky on GitHub (Dec 20, 2024). Original GitHub issue: https://github.com/Mintplex-Labs/anything-llm/issues/2883 ### What would you like to see? ## Description Request to integrate Model Context Protocol (MCP) support into AnythingLLM to enhance interoperability and standardization of context handling across different LLM implementations. ## Motivation MCP is becoming a standard protocol for AI applications, offering a "USB-C port for AI applications" that would benefit AnythingLLM by: - Providing standardized context handling across different LLM providers - Enabling easier integration with other MCP-compatible tools and services - Improving the flexibility of switching between different LLM providers - Adding support for standardized prompt templates and tool interactions ## Proposed Implementation The implementation would require: ### 1. Core MCP Features - Resource handling for data access and context management - Prompt template support with argument handling - Tool integration capabilities - Support for both synchronous and asynchronous operations ### 2. Integration Points - MCP client implementation for connecting to MCP servers - Server configuration support - Protocol handlers for MCP communication - Resource update subscription mechanisms ### 3. Security Considerations - Proper authentication and authorization - Secure handling of API keys and sensitive information - Data security best practices ## Benefits 1. **Standardization**: Align with emerging industry standards for LLM context handling 2. **Interoperability**: Better integration with other AI tools and services 3. **Flexibility**: Easier switching between different LLM providers 4. **Extensibility**: Simplified addition of new features and capabilities 5. **Community**: Access to growing ecosystem of MCP-compatible tools ## Technical Requirements - Implementation of MCP client capabilities - Support for MCP protocol specifications - Integration with existing AnythingLLM architecture - Backward compatibility with current features ## Additional Context MCP is being adopted by various applications including: - Claude Desktop - Sourcegraph Cody - Continue - Zed - Firebase Genkit
yindo added the enhancementfeature request labels 2026-02-22 18:26:44 -05:00
yindo closed this issue 2026-02-22 18:26:44 -05:00
yindo commented 2026-02-22 18:26:45 -05:00
Author
Owner
Copy Link
Copy Source

@gschoepp commented on GitHub (Dec 22, 2024):

I was tempted to have Claude write a heartfelt support missive to put here, but nah. MCP would be an amazing add to AnythingLLM. MCP is basically the AnythingLLM of the function calling space... pretty please? :) Peace.

@gschoepp commented on GitHub (Dec 22, 2024): I was tempted to have Claude write a heartfelt support missive to put here, but nah. MCP would be an **amazing** add to AnythingLLM. MCP is basically the AnythingLLM of the function calling space... pretty please? :) Peace.
yindo commented 2026-02-22 18:26:45 -05:00
Author
Owner
Copy Link
Copy Source

@spencerthayer commented on GitHub (Dec 23, 2024):

I recommend starting by connecting AnythingLLM to MCP via a custom agent skill plugin: ' agent-skills/mcp-connector/`. This isn't working quite right for me but maybe you can get it to work.

// plugin.json
{
  "active": true,
  "hubId": "mcp-connector",
  "name": "MCP Protocol Connector", 
  "schema": "skill-1.0.0",
  "version": "1.0.0",
  "description": "Connects to MCP-compatible tools and APIs",
  "author": "If he makes it work, @GitDakky",
  "license": "MIT",
  
  "setup_args": {
    "MCP_HOST": {
      "type": "string",
      "required": true,
      "input": {
        "type": "text",
        "default": "localhost:3000",
        "placeholder": "localhost:3000",
        "hint": "Address of your MCP server"
      }
    }
  },

  "entrypoint": {
    "file": "handler.js",
    "params": {
      "tool_name": {
        "description": "Name of the MCP tool to call",
        "type": "string"
      },
      "args": {
        "description": "Arguments for the tool",
        "type": "object"
      }
    }
  },

  "examples": [
    {
      "prompt": "Calculate 5 plus 3",
      "call": "{\"tool_name\": \"calculate_sum\", \"args\": {\"a\": 5, \"b\": 3}}"
    },
    {
      "prompt": "Get slideshow data",
      "call": "{\"tool_name\": \"httpbin_json\", \"args\": {}}"
    }
  ]
}

// handler.js
module.exports.runtime = {
  handler: async function ({ tool_name, args }) {
    try {
      // Log the attempt
      this.logger(`Attempting to call MCP tool: ${tool_name}`);
      this.introspect(`Connecting to MCP server at ${this.runtimeArgs.MCP_HOST}`);

      // Make the MCP tool call
      const response = await fetch(`http://${this.runtimeArgs.MCP_HOST}/tools/${tool_name}`, {
        method: 'POST',
        headers: {
          'Content-Type': 'application/json'
        },
        body: JSON.stringify(args)
      });

      if (!response.ok) {
        throw new Error(`MCP tool call failed: ${response.statusText}`);
      }

      const result = await response.json();
      return JSON.stringify(result);

    } catch (error) {
      this.logger(`Error in MCP connector: ${error.message}`);
      return `Failed to execute MCP tool: ${error.message}`;
    }
  }
};

# README.md
# MCP Protocol Connector

This custom agent skill allows AnythingLLM to interact with MCP-compatible tools and services.

## Setup
1. Install the skill in your AnythingLLM plugins directory
2. Configure the MCP_HOST in the agent skills settings
3. Ensure your MCP server is running and accessible

## Usage
The skill can call any tool registered with your MCP server:

Calculate sum example:
{
  "tool_name": "calculate_sum",
  "args": {
    "a": 5,
    "b": 3
  }
}

Get external data example:
{
  "tool_name": "httpbin_json",
  "args": {}
}

## Requirements
- AnythingLLM running in supported environment
- Access to an MCP-compatible server

## Error Handling
The skill provides detailed error messages and logging for:
- Connection failures
- Invalid tool names
- Invalid arguments
- Server errors

## Logging
All tool calls are logged with:
- Tool name
- Arguments
- Success/failure status
- Error messages (if any)

## Support
If you encounter any issues:
1. Check the AnythingLLM logs
2. Verify your MCP server is running
3. Ensure the MCP_HOST configuration is correct

Installation steps:

  1. Create the directory:
mkdir -p plugins/agent-skills/mcp-connector
  1. Copy each file into this directory:
# From the mcp-connector directory
touch plugin.json handler.js README.md
# Then paste the contents into each file
  1. Set permissions:
chmod 644 plugin.json README.md
chmod 755 handler.js
  1. Restart AnythingLLM to load the new agent skill

The skill will then be available in your AnythingLLM instance and can be configured through the UI.

@spencerthayer commented on GitHub (Dec 23, 2024): I recommend starting by connecting AnythingLLM to MCP via a custom agent skill plugin: ' agent-skills/mcp-connector/`. This isn't working quite right for me but maybe you can get it to work. ```json // plugin.json { "active": true, "hubId": "mcp-connector", "name": "MCP Protocol Connector", "schema": "skill-1.0.0", "version": "1.0.0", "description": "Connects to MCP-compatible tools and APIs", "author": "If he makes it work, @GitDakky", "license": "MIT", "setup_args": { "MCP_HOST": { "type": "string", "required": true, "input": { "type": "text", "default": "localhost:3000", "placeholder": "localhost:3000", "hint": "Address of your MCP server" } } }, "entrypoint": { "file": "handler.js", "params": { "tool_name": { "description": "Name of the MCP tool to call", "type": "string" }, "args": { "description": "Arguments for the tool", "type": "object" } } }, "examples": [ { "prompt": "Calculate 5 plus 3", "call": "{\"tool_name\": \"calculate_sum\", \"args\": {\"a\": 5, \"b\": 3}}" }, { "prompt": "Get slideshow data", "call": "{\"tool_name\": \"httpbin_json\", \"args\": {}}" } ] } ``` ```javascript // handler.js module.exports.runtime = { handler: async function ({ tool_name, args }) { try { // Log the attempt this.logger(`Attempting to call MCP tool: ${tool_name}`); this.introspect(`Connecting to MCP server at ${this.runtimeArgs.MCP_HOST}`); // Make the MCP tool call const response = await fetch(`http://${this.runtimeArgs.MCP_HOST}/tools/${tool_name}`, { method: 'POST', headers: { 'Content-Type': 'application/json' }, body: JSON.stringify(args) }); if (!response.ok) { throw new Error(`MCP tool call failed: ${response.statusText}`); } const result = await response.json(); return JSON.stringify(result); } catch (error) { this.logger(`Error in MCP connector: ${error.message}`); return `Failed to execute MCP tool: ${error.message}`; } } }; ``` ```markdown # README.md # MCP Protocol Connector This custom agent skill allows AnythingLLM to interact with MCP-compatible tools and services. ## Setup 1. Install the skill in your AnythingLLM plugins directory 2. Configure the MCP_HOST in the agent skills settings 3. Ensure your MCP server is running and accessible ## Usage The skill can call any tool registered with your MCP server: Calculate sum example: { "tool_name": "calculate_sum", "args": { "a": 5, "b": 3 } } Get external data example: { "tool_name": "httpbin_json", "args": {} } ## Requirements - AnythingLLM running in supported environment - Access to an MCP-compatible server ## Error Handling The skill provides detailed error messages and logging for: - Connection failures - Invalid tool names - Invalid arguments - Server errors ## Logging All tool calls are logged with: - Tool name - Arguments - Success/failure status - Error messages (if any) ## Support If you encounter any issues: 1. Check the AnythingLLM logs 2. Verify your MCP server is running 3. Ensure the MCP_HOST configuration is correct ``` Installation steps: 1. Create the directory: ```bash mkdir -p plugins/agent-skills/mcp-connector ``` 2. Copy each file into this directory: ```bash # From the mcp-connector directory touch plugin.json handler.js README.md # Then paste the contents into each file ``` 3. Set permissions: ```bash chmod 644 plugin.json README.md chmod 755 handler.js ``` 4. Restart AnythingLLM to load the new agent skill The skill will then be available in your AnythingLLM instance and can be configured through the UI.
yindo commented 2026-02-22 18:26:45 -05:00
Author
Owner
Copy Link
Copy Source

@timothycarambat commented on GitHub (Dec 28, 2024):

Indeed this would be better suited as an agent skill, certainly at least in the short term. It would suffice for your flexibility and needs as well require zero lift from us to do. If it become the "defacto" standard among the LLM API layer, then we would then make it core.

@timothycarambat commented on GitHub (Dec 28, 2024): Indeed this would be better suited as an agent skill, certainly at least in the short term. It would suffice for your flexibility and needs as well require zero lift from us to do. If it become the "defacto" standard among the LLM API layer, then we would then make it core.
yindo commented 2026-02-22 18:26:45 -05:00
Author
Owner
Copy Link
Copy Source

@evangineer commented on GitHub (Jan 25, 2025):

I recommend starting by connecting AnythingLLM to MCP via a custom agent skill plugin: ' agent-skills/mcp-connector/`. This isn't working quite right for me but maybe you can get it to work.

(truncated)

@spencerthayer I'm wondering what the issue was with getting the MCP agent skill working and if you have a solution for it.

@evangineer commented on GitHub (Jan 25, 2025): > I recommend starting by connecting AnythingLLM to MCP via a custom agent skill plugin: ' agent-skills/mcp-connector/`. This isn't working quite right for me but maybe you can get it to work. > > (truncated) @spencerthayer I'm wondering what the issue was with getting the MCP agent skill working and if you have a solution for it.
yindo commented 2026-02-22 18:26:45 -05:00
Author
Owner
Copy Link
Copy Source

@computersrmyfriends commented on GitHub (Feb 1, 2025):

Did anyone get the mcp skill to work? MCP is the defacto as of today, supported by several agentic frameworks, coding ides and ollama

@computersrmyfriends commented on GitHub (Feb 1, 2025): Did anyone get the mcp skill to work? MCP is the defacto as of today, supported by several agentic frameworks, coding ides and ollama
yindo commented 2026-02-22 18:26:45 -05:00
Author
Owner
Copy Link
Copy Source

@spencerthayer commented on GitHub (Feb 1, 2025):

@computersrmyfriends MCP is not the “defacto” at all. The only working MCP client right now is Claude Desktop. Anthropics competitors will not want to adopt standardization until it is clear who has majority market share. And right now that’s no one. This should remain an agent.

@spencerthayer commented on GitHub (Feb 1, 2025): @computersrmyfriends MCP is not the “defacto” at all. The only working MCP client right now is Claude Desktop. Anthropics competitors will not want to adopt standardization until it is clear who has majority market share. And right now that’s no one. This should remain an agent.
yindo commented 2026-02-22 18:26:45 -05:00
Author
Owner
Copy Link
Copy Source

@timothycarambat commented on GitHub (Feb 1, 2025):

@spencerthayer there is another thread on this that I decided to reopen. But I agree, it is indeed not defacto, but there is no need to have MCP work be rehashed or rebuilt in our tooling if and only if we can easily "wrap" it in our framework.

The push by Anthropic and its partners and co-marketing certainly has made a wave in agent coding. How ephemeral that is will be determined, which is why we are not going to uproot our entire extremely flexible framework into a third party maintained one.

If we can unlock a way to be able to facilitate a low-friction import-and-run-mcp style of UX in AnythingLLM, I am happy to explore that kind of flow. Just like how we don't force an LLM, we shouldn't force our framework either and without a doubt some people are going to adopt it.

It would be a high effort and low reward undertaking to do this if the nature of MCP is short-lived.

Conversation moving to https://github.com/Mintplex-Labs/anything-llm/issues/3000

@timothycarambat commented on GitHub (Feb 1, 2025): @spencerthayer there is another thread on this that I decided to reopen. But I agree, it is indeed _not defacto_, but there is no need to have MCP work be rehashed or rebuilt in our tooling _if and only if_ we can easily "wrap" it in our framework. The push by Anthropic and its partners and co-marketing certainly has made a wave in agent coding. How ephemeral that is will be determined, which is why we are not going to uproot our entire _extremely flexible_ framework into a third party maintained one. If we can unlock a way to be able to facilitate a low-friction import-and-run-mcp style of UX in AnythingLLM, I am happy to explore that kind of flow. Just like how we don't force an LLM, we shouldn't force our framework either and without a doubt _some people_ are going to adopt it. It would be a high effort and low reward undertaking to do this if the nature of MCP is short-lived. Conversation moving to https://github.com/Mintplex-Labs/anything-llm/issues/3000
yindo changed title from [FEAT]: Model Context Protocol (MCP) Integration to [GH-ISSUE #2883] [FEAT]: Model Context Protocol (MCP) Integration 2026-06-05 14:42:57 -04:00
Sign in to join this conversation.
No Branch/Tag Specified
Branches Tags
Labels
Clear labels
Desktop
Docker
Integration Request
Integration Request
OS: Linux
OS: Mobile
OS: Windows
UI/UX
blocked
bug
bug
core-team-only
documentation
duplicate
embed-widget
enhancement
feature request
github_actions
good first issue
investigating
needs info / can't replicate
possible bug
pull-request
Mirrored from GitHub Pull Request
 question
stage: specifications
wontfix
No Label enhancement feature request
Milestone
No items
No Milestone
Projects
Clear projects
No project
Assignees
Clear assignees
yindo
No Assignees
1 Participants
Notifications
Due Date
No due date set.
Dependencies

No dependencies set.

Reference: Mintplex-Labs/anything-llm#1834
Delete Branch "%!s()"

Deleting a branch is permanent. Although the deleted branch may continue to exist for a short time before it actually gets removed, it CANNOT be undone in most cases. Continue?