Documentation Index Fetch the complete documentation index at: https://docs.cubic.dev/llms.txt
Use this file to discover all available pages before exploring further.
Connect cubic’s MCP server to your coding agent so it can read PR review issues, use codebase wikis, inspect scan results, triage scan issues, and read team review learnings while you work.
Use the exact endpoint https://www.cubic.dev/api/mcp. Some OAuth clients reject
https://cubic.dev/api/mcp because the protected resource origin does not match.
Cursor
Cursor Agent
Claude Code
Codex
Gemini CLI
OpenCode
Droid
Pi
VS Code
Open settings
Open Cursor Settings .
Open Tools and MCP
Go to Tools and MCP .
Choose MCP section
In MCP, use Home MCP Servers or your project-specific MCP section.
Create a server
Click New MCP server .
Save the config
Add this to the config and save: "cubic" : { "type" : "http" , "url" : "https://www.cubic.dev/api/mcp" }
Add the MCP server
Add this to ~/.cursor/mcp.json or project .cursor/mcp.json: { "mcpServers" : { "cubic" : { "type" : "http" , "url" : "https://www.cubic.dev/api/mcp" } } }
Log in
Run these commands from the project where you use Cursor Agent: cursor-agent mcp enable cubic cursor-agent mcp login cubic
Verify
cursor-agent mcp list-tools cubic
Add the MCP server
claude mcp add --transport http --scope user cubic https://www.cubic.dev/api/mcp
Log in
Open Claude Code and run: Choose cubic and complete the browser OAuth flow.
Optional skills and slash commands
Install cubic’s Claude Code skills separately: /plugin marketplace add mrge-io/skills /plugin install cubic@cubic
Add the MCP server
codex mcp add cubic --url https://www.cubic.dev/api/mcp
Add the MCP server
gemini mcp add --transport http --scope user cubic https://www.cubic.dev/api/mcp
Add the MCP server
Add this to ~/.config/opencode/opencode.json: { "$schema" : "https://opencode.ai/config.json" , "mcp" : { "cubic" : { "type" : "remote" , "url" : "https://www.cubic.dev/api/mcp" , "enabled" : true } } }
Add the MCP server
droid mcp add cubic https://www.cubic.dev/api/mcp --type http
Log in
Open Droid, run /mcp, choose cubic, and complete the browser OAuth flow.
Verify
The /mcp panel should show cubic as connected.
Install cubic's Pi target
npx @cubic-plugin/cubic-plugin install --to pi
The installer writes Pi’s MCP config to ~/.config/mcp/mcp.json.
Install the MCP adapter (if needed)
pi install npm:pi-mcp-adapter
Skip this step if pi-mcp-adapter is already installed.
Initialize and verify the MCP server
Run Pi’s guided setup: Confirm the MCP config contains: { "mcpServers" : { "cubic" : { "auth" : "oauth" , "url" : "https://www.cubic.dev/api/mcp" } } }
Add the MCP server
Open the Command Palette, run MCP: Add Server , choose HTTP, and enter: https://www.cubic.dev/api/mcp
Log in
Follow VS Code’s OAuth prompt for the cubic MCP server.
Verify
Run MCP: List Servers and confirm cubic is running.
What you can ask
“show cubic issues on this PR”
“list wiki pages for this repo”
“what review learnings apply here?”
“show scan findings with severity at least 7”
Skills and slash commands MCP gives your agent access to cubic tools. For agents that also support skills or slash commands, install those separately:
npx @cubic-plugin/cubic-plugin install --skills-only
Command Description /cubic-run-review or /cubic:run-reviewRun a local cubic AI code review from your editor. Requires the cubic CLI.
The /cubic-run-review and /cubic:run-review commands require the cubic CLI. See Local CLI review
for installation. 