- IDENTITY
- SECURITY
- ACCESS
- REVIEW
Short answer
How can I add new agents to my control plane?
You tell your AI to register a new agent and define its security boundaries. Your agent uses Runlayer to create the identity and attach the necessary policies and MCPs. You get a fully configured agent ready for deployment.
The onboarding outcomes.
Where every agent lands.
Your AI manages the setup process from registration to permissioning.
IDENTITY
New agent created
The AI registers a unique identity for the agent in your control plane. You get a new, trackable entity in your system.
SECURITY
Policies applied
The AI attaches specific security rules to the new agent. This ensures the agent only touches what it is allowed to.
ACCESS
MCPs assigned
The AI links the agent to the required Model Context Protocol tools. This gives the agent the specific capabilities it needs to work.
REVIEW
Human oversight
The AI presents the configuration for your final approval. You decide if the permissions and tools match your requirements before finishing.
The workflow
What your AI does when the request arrives.
Your AI acts as the administrator for your agent fleet.
Register the agent
The AI creates a new agent entry in your control plane with the name and parameters you provide.
create_agentDefine security
The AI sets up the specific security constraints that govern the agent's behavior.
create_policyVerify existing agents
The AI checks your current agent list to ensure there are no naming conflicts or duplicate identities.
list_agentsFinalize setup
The AI confirms the agent is registered and ready to receive further instructions.
create_agent
Try it
Copy these to start.
Use these prompts to kick off the onboarding process.
Starting points
Replace the names in brackets with your actual agent names or policy IDs.
Runlayer Connector
You're all set. Choose your MCP client and follow the setup instructions.
https://edge.vinkius.com/vk_preview_PaTg4jiaCBAOZI6pcScShh9467XvfcKP4495NMGs/mcpClaude Desktop
Follow the steps below to connect in seconds.
- 1In Claude Desktop, open Settings → Connectors.
- 2Click “Add custom connector” and paste the connector link above as the remote MCP server URL.
- 3Click Add and start a new chat — Runlayer capabilities are ready to use.
{
"mcpServers": {
"runlayer-mcp": {
"url": "https://edge.vinkius.com/vk_preview_PaTg4jiaCBAOZI6pcScShh9467XvfcKP4495NMGs/mcp"
}
}
}Create a new agent named 'DataScraper-01' and give it a standard read-only policy.
Register an agent called 'CustomerSupport-Bot' and assign it the 'web-search' MCP.
List all my current agents to see who is already registered.
Set up a new agent named 'Finance-Analyst' with a strict security policy.
Claude
ChatGPT
Cursor
VS Code
Windsurf
Claude Code
JetBrains
Cline
Start here
Connect Runlayer once, then ask.
Connect your account via the Vinkius dashboard. Your credentials stay encrypted, and you can immediately start managing agents through your preferred AI client.
Connect Runlayer to your AIFAQ
How this task behaves.
- 01
Can the AI delete existing agents?
No. The current tools only allow for creating agents, creating policies, and listing agents. Deletion must be done manually.
- 02
Does the AI decide which security policies to use?
No. You must tell the AI which specific policy to apply to the new agent.
- 03
Can I see a list of all my agents?
Yes. Your AI can pull a full list of registered agents at any time.
- 04
Will the AI change my existing security policies?
No. The AI can create new policies, but it cannot modify or edit policies that already exist.
- 05
Can I assign multiple MCPs to one agent?
Yes. You can instruct the AI to assign the necessary MCPs during the setup process.
More questions about Runlayer? The Connector page answers them. See everything the Runlayer Connector can do
Connect Runlayer to Claude, Cursor, ChatGPT & more
