DevRev MCP Server
# DevRev MCP server
## Overview
A Model Context Protocol server for DevRev. It is used to search and retrieve information using the DevRev APIs.
## Tools
- `search`: Search for information using the DevRev search API with the provided query and namespace.
- `get_object`: Get all information about a DevRev object using its ID.
## Configuration
### Get the DevRev API key
1. Go to https://app.devrev.ai/signup and create an account.
2. Import your data from your existing data sources like Salesforce, Zendesk while following the instructions [here](https://devrev.ai/docs/import#available-sources).
3. Generate an access token while following the instructions [here](https://developer.devrev.ai/public/about/authentication#personal-access-token-usage).
### Usage with Claude Desktop
On MacOS: `~/Library/Application\ Support/Claude/claude_desktop_config.json`
On Windows: `%APPDATA%/Claude/claude_desktop_config.json`
<details>
<summary>Published Servers Configuration</summary>
```
"mcpServers": {
"devrev": {
"command": "uvx",
"args": [
"devrev-mcp"
],
"env": {
"DEVREV_API_KEY": "YOUR_DEVREV_API_KEY"
}
}
}
```
</details>
<details>
<summary>Development/Unpublished Servers Configuration</summary>
```
"mcpServers": {
"devrev": {
"command": "uv",
"args": [
"--directory",
"Path to src/devrev_mcp directory",
"run",
"devrev-mcp"
],
"env": {
"DEVREV_API_KEY": "YOUR_DEVREV_API_KEY"
}
}
}
```
</details>
TDQS
Scored across 2 tools
The two tools have clearly distinct purposes: get_object retrieves a specific object by ID, while search performs a broader query-based lookup. There is no overlap or ambiguity between these operations.
Both tools follow a consistent verb-based naming pattern (get_ and search), with clear, descriptive names that align well with their functions. No deviations or mixed conventions are present.
With only two tools, the server feels severely under-scoped for a DevRev platform, which typically involves objects like issues, tickets, or users. This minimal set lacks essential CRUD operations (e.g., create, update, delete) and limits agent functionality.
The tool surface is severely incomplete for a DevRev domain, offering only retrieval and search capabilities. There are no tools for creating, updating, or deleting objects, which are fundamental operations in such systems, leading to significant gaps in agent workflows.