# sequa-ai/sequa-mcp [Health: Active]

**Category:** 🤖 Coding Agents  
**Repository:** https://github.com/sequa-ai/sequa-mcp  
**GitHub Stars:** 22  
**npm Downloads (last month):** 162  
**Views:** 5  
**Installs:** 0  
**Upvotes:** 0  
**Directory Page:** https://allmcps.com/mcp/sequa-ai-sequa-mcp

## Description
Stop stitching context for Copilot and Cursor. With Sequa MCP, your AI tools know your entire codebase and docs out of the box.

## Claude Desktop Quick Installation
Install path detected from listing signals. Uses `npx` (confidence: high):

```json
"mcpServers": {
  "sequa-mcp": {
    "command": "npx",
    "args": ["-y","@sequa-ai/sequa-mcp@latest"]
  }
}
```

## Documentation

## What sequa-ai/sequa-mcp MCP server does

The sequa-ai/sequa-mcp MCP server exposes Sequa project knowledge to AI development clients through the Model Context Protocol. Sequa is described as a contextual knowledge engine that builds and maintains documentation from a codebase, including architecture overviews, API documentation, runbooks, and onboarding guides.

The available context can span repositories and documentation, along with tickets and internal standards. Developers can ask questions in plain language and receive answers with source links. The intended use is to give coding assistants current project information while they generate or modify code, reducing the need to assemble repository and documentation context manually.

## How it works

The package acts as a proxy to a Sequa MCP endpoint. Its NPX command takes a setup URL as an argument:

```bash
npx -y @sequa-ai/sequa-mcp@latest https://mcp.sequa.ai/v1/setup-code-assistant
```

A project-specific URL can replace the example endpoint. Clients that support remote MCP URLs can use the endpoint directly; clients configured with a command can launch the NPX package and pass the URL as an argument.

Before connecting, create or sign in to a Sequa account and set up a project in the Sequa application. The project's MCP Setup URLs provide the transport-specific URL or configuration. Without a project, the proxy may launch but connections will return authentication or project errors.

## Setup and configuration

The sequa-ai/sequa-mcp MCP server can be configured in Cursor, Claude Desktop, Windsurf, VS Code, Cline, Highlight AI, and Augment Code. Configuration varies by client. Cursor uses a server entry with a `url`, while the other examples generally invoke `npx` with the package name and setup URL.

For Claude Desktop, the configuration is added under Settings → Developer → Edit Config. Cursor reads `~/.cursor/mcp.json`; Windsurf reads `~/.codeium/windsurf/mcp_config.json`; VS Code uses `.vscode/mcp.json`; and Cline uses `cline_mcp_settings.json`. Highlight AI accepts the NPX command through its custom-plugin interface, and Augment Code supports either the command or an `augment_config.json` entry.

The setup URL is the key configuration value. Use the URL generated for the intended Sequa project rather than assuming the example endpoint applies to every project. No environment variables are specified in the provided setup instructions.

## Tools and capabilities

The documented capabilities are knowledge-oriented rather than a named list of individual MCP tools. They include access to:

- Codebase and repository context
- Generated documentation such as architecture, API, runbook, and onboarding material
- Documentation collected across repositories
- Tickets and internal engineering standards
- Plain-language questions with source links in the answers

Sequa is intended to keep documentation aligned with changes in the underlying code. The README presents this as continuous updating from code, so assistants can use a maintained knowledge base instead of relying only on static documents.

## Limitations and notes

The repository material does not provide a detailed tool schema, request examples, permission model, rate limits, pricing information, or supported operating systems. It also does not list environment variables or a separate local indexing process.

A Sequa account and project are prerequisites. The server is not expected to provide a usable connection from the setup command alone: the README specifically notes that omitting project creation results in authentication or project errors. Client compatibility depends on MCP support and the transport available from the project's MCP Setup URLs.

_Full upstream README: https://allmcps.com/mcp/sequa-ai-sequa-mcp/readme_