For AI agents: a documentation index is available at the root level at /llms.txt and /llms-full.txt. Append /llms.txt to any URL for a page-level index, or .md for the markdown version of any page.
Logo
Resources
Log inGet a demo
GuidesModelsAPI reference
GuidesModelsAPI reference
    • Get started
    • Install skills
  • Features
    • Projects
    • Management API keys
    • Cost governance and savings
    • Tool calling
    • Web search
    • Context compression
      • Overview
      • Cursor
      • Zed
      • Continue.dev
      • OpenCode
      • Claude Code
      • Codex
  • Security & Compliance
    • Customer blocklist
    • Geo-location routing
    • Prompt injection protection
    • Data loss prevention
    • Audit trail
    • Roles and permissions
    • Zero data retention
    • Provider terms

Get started

  • Overview
  • Introduction
  • Unified API
  • Linked Account
  • Merge Link
  • Use cases

Implementation

  • Sandboxes
  • SDKs
  • API access
  • Syncing data
  • Writing data
  • Data minimization
  • Supplemental data
  • Errors
  • Integration metadata

API reference

  • ATS
  • HRIS
  • Accounting
  • Ticketing
  • CRM
  • File Storage
  • Knowledge Base
  • Chat

Resources

  • Help Center
  • Merge.dev
  • Changelog
© Merge 2026Terms of usePrivacy policy
UnifiedAgent HandlerGateway
UnifiedAgent HandlerGateway
Resources
Log inGet a demo
On this page
  • The three inputs
  • Supported IDEs
  • Next steps
FeaturesBuilding with coding agents

Building with coding agents

Route your IDE or terminal coding agent through Merge Gateway with a single API key
Was this page helpful?
Previous

Context compression

Next

Cursor

Most coding tools, from OpenAI-compatible IDEs like Cursor and Zed to terminal agents like OpenCode, Claude Code, and Codex, let you plug in a custom provider. Point that provider at Merge Gateway and you get every model Gateway supports, including OpenAI, Anthropic, Google, and Kimi, through a single API key. Routing policies, cost governance, and full request observability come built in, visible in the Gateway dashboard.

The three inputs

The OpenAI-compatible IDE guides on this page (Cursor, Zed, Continue.dev) all ask for the same three values:

InputValue
Base URLhttps://api-gateway.merge.dev/v1/openai
API keyYour key from gateway.merge.dev/settings/api-keys
Model nameprovider/model format. For example, anthropic/claude-opus-4-6, openai/gpt-5.2, or google/gemini-2.5-pro.

Get the full list of available models from GET /v1/models or the models page in the dashboard.

OpenCode, Claude Code, and Codex follow different setup shapes (a custom provider in opencode.json, an ANTHROPIC_BASE_URL override, and a model provider + profile in ~/.codex/config.toml respectively); see their dedicated guides below for the exact config.

Supported IDEs

Cursor

Override OpenAI Base URL in Cursor Settings. Supports Ask and Plan modes.

Zed

Add Gateway as an OpenAI-compatible provider in settings.json

Continue.dev

One-entry config.yaml change. Works in VS Code and JetBrains.

OpenCode

Add Merge Gateway as a custom provider in opencode.json. Works for tool calls and file edits.

Claude Code

Override ANTHROPIC_BASE_URL and auth token. Works for tool calls and file edits.

Codex

Register Merge Gateway in ~/.codex/config.toml and opt in via codex --profile merge_gateway

Next steps

Projects

Use per-project API keys so IDE usage is tracked and billed separately

Cost governance

Set spend limits and model allow-lists for IDE traffic