
Connect to JetBrains
IntelliJ, PyCharm, WebStorm, Rider — MCP through AI Assistant. One configuration works for all of them.
Which MCP will you connect? The commands below change according to your choice.
- 1
Check the version and AI Assistant
MCP support arrived in the 2025.1 line of IDEs, with the AI Assistant plugin (or Junie) installed and enabled. Under Help → About, check the IDE version.
- 2
Generate your API key
Sign in to app.newalgorithm.io with your account, open Keys & apps and generate a key. It is shown only once: copy it and store it somewhere safe.
Key formatna_live_xxxxxxxxxxxxxxxxx - 3
Open the MCP configuration
Settings/Preferences → Tools → AI Assistant → Model Context Protocol (MCP). Use the add button and choose the JSON entry.
- 4
Declare the NewAlgorithm server
The format is the same mcpServers used by the other clients:
AI Assistant MCP configuration{ "mcpServers": { "NewAlgorithm-TOTVS": { "url": "https://totvs.mcp.newalgorithm.io/mcp", "headers": { "Authorization": "Bearer na_live_xxxxxxxxxxxxxxxxx" } } } } - 5
If your version only accepts a local command
Older versions of the plugin only speak stdio. The bridge solves it, and requires Node.js:
Bridge alternative{ "mcpServers": { "NewAlgorithm-TOTVS": { "command": "npx", "args": [ "-y", "mcp-remote", "https://totvs.mcp.newalgorithm.io/mcp", "--header", "Authorization: Bearer na_live_xxxxxxxxxxxxxxxxx" ] } } } - 6
Apply and confirm
Click Apply. The server appears in the list as connected; in the AI Assistant chat, the tool picker starts listing the na_* tools.
Expected: na_prepare, na_search, na_route, na_specialists, na_version
The configuration lives at the IDE level, not the project level — it applies to every repository you open in that IDE. Each JetBrains product keeps its own, so repeat it in each IDE you use.
Authentication and tool details in the Docs.