Two kinds of integrations live here. AI assistants and coding agents connect over MCP and read your workflows, requirements and domain model straight from your account. Platform integrations run inside Qlerify itself: Jira, Azure DevOps and GitHub.
Coding agents and IDEs. Generate a key in Qlerify, paste it into the client’s config.
Chat assistants. No key at all — you add a custom connector and sign in.
Qlerify runs an MCP server. Connect a client once and it can read your workflows, requirements and domain model straight from your account. Pick your client from the list on the left.
If your client authenticates with an API key, generate one first on the Integrations page in your Qlerify account, then paste it wherever the steps below say YOUR_API_KEY. OAuth clients need no key.
Run this in your project’s folder:
claude mcp add --transport http qlerify https://mcp.qlerify.com \ --header "x-api-key: YOUR_API_KEY"
Replace YOUR_API_KEY with the key you generated in Qlerify. This adds Qlerify to that project only. To use it in every project, add --scope user to the command.
Adds the Qlerify plugin, with its skills, to Claude Code:
/plugin marketplace add qlerify/qlerify-plugins /plugin install qlerify@qlerify-plugins
The plugin does not update by itself. To get new versions automatically, run /plugin, open Marketplaces, select qlerify-plugins and choose Enable auto-update.
Add this to your Cursor MCP settings file, ~/.cursor/mcp.json:
{
"mcpServers": {
"qlerify": {
"url": "https://mcp.qlerify.com",
"headers": {
"x-api-key": "YOUR_API_KEY"
}
}
}
}Replace YOUR_API_KEY with the key you generated in Qlerify.
Clone the plugins repository, or use GitHub’s Code → Download ZIP:
git clone https://github.com/qlerify/qlerify-plugins.git
Copy the skill folders from plugins/qlerify/skills/ into one of these:
.cursor/
└── skills/
├── workflow-creation/
│ ├── SKILL.md
│ ├── references/
│ └── examples/
└── code-generation/
└── SKILL.mdReload the window afterwards (Cmd+Shift+P → Developer: Reload Window) so Cursor picks the skills up.
Cursor’s own skills documentation covers the details.
Paste this into the config, merging into mcpServers if you already have others:
{
"mcpServers": {
"qlerify": {
"serverUrl": "https://mcp.qlerify.com",
"headers": {
"Authorization": "Bearer YOUR_API_KEY",
"Content-Type": "application/json"
}
}
}
}Clone the plugins repository, or use GitHub’s Code → Download ZIP:
git clone https://github.com/qlerify/qlerify-plugins.git
Copy the skill folders from plugins/qlerify/skills/ into one of these:
.agents/
└── skills/
├── workflow-creation/
│ ├── SKILL.md
│ ├── references/
│ └── examples/
└── code-generation/
└── SKILL.mdThe full set of skills lives in the qlerify-plugins repository.
Antigravity’s own skills documentation covers the details.

Add this to your MCP config file, .vscode/mcp.json:
{
"servers": {
"qlerify": {
"type": "http",
"url": "https://mcp.qlerify.com",
"headers": {
"x-api-key": "YOUR_API_KEY"
}
}
}
}Replace YOUR_API_KEY with the key you generated in Qlerify.
Adds the Qlerify plugin, with its skills, to VS Code:
VS Code asks to add the plugin marketplace. Click Add Marketplace, then click Install on Qlerify in the Extensions view that opens.
Or add the marketplace yourself in your user settings (Preferences: Open User Settings (JSON)):
"chat.plugins.enabled": true, "chat.plugins.marketplaces": [ "github/copilot-plugins", "github/awesome-copilot#marketplace", "qlerify/qlerify-plugins" ]
Then open the Extensions view, search for @agentPlugins and install Qlerify.
This setting replaces VS Code’s default list, so the snippet also lists VS Code’s own marketplaces. The plugin’s skills then show up in the chat’s Configure Skills menu.
Run this in your terminal:
gemini mcp add \ --scope user \ --transport http \ --header "x-api-key: YOUR_API_KEY" \ qlerify "https://mcp.qlerify.com"
Replace YOUR_API_KEY with the key you generated in Qlerify.
One command installs all the Qlerify skills:
gemini skills install https://github.com/qlerify/qlerify-plugins.git \ --path plugins/qlerify/skills
Sign-in happens over OAuth, so there is no API key to generate.
Opens claude.ai with the name and URL filled in. Click Add, then sign in with your Qlerify account.
Or add it yourself:
Adds the Qlerify skills to Claude:
ChatGPT hides custom MCP apps until Developer mode is on. Find it under Settings → Security and login.
The integrations that run inside Qlerify rather than through an AI client.
Export cards from a Qlerify workflow to Jira Cloud as issues. Each linked card then shows its Jira issue inside Qlerify: issue key, summary, type, assignee and status.
Export workflow data from Qlerify to Azure Boards.
Push the domain model specification (JSON) of a bounded context to a repository and branch of your choice, from the Push Spec tab on the Domain Model page. Your coding agent then builds from that file.
Before pushing, Qlerify validates the model, shows the exact file that will be written, and shows a diff if a version is already in the repository.