- USAGE
- ACTIVITY
- CLIENTS
- HUMAN
Short answer
How can I audit AI Bridge usage?
Your AI client pulls real-time data on active sessions, model types, and connected clients from your remote infrastructure. You get a clear view of how your development environment is being used. You'll see the specific breakdown of these metrics below.
Audit outcomes
Where your data lands.
The AI organizes your infrastructure data into these specific views.
USAGE
Model distribution
The AI identifies which models are currently active. You get a list of model types running in your environment.
ACTIVITY
Active session tracking
Your agent pulls a list of all ongoing sessions. You can see exactly which sessions are live.
CLIENTS
Client identification
The AI lists every client connected to the bridge. This shows you which tools are driving the most traffic.
HUMAN
Manual review needed
The AI provides the raw session data. You decide which specific threads require a deeper look by a person.
The workflow
What your AI does when the request arrives.
The AI uses the Coder (Remote Dev) MCP to pull data from your remote environment.
Identify active users
The AI finds all currently running AI sessions to establish a baseline of activity.
list_ai_sessionsMap model types
It checks which specific models are being called to help you understand resource allocation.
list_ai_modelsCheck connection points
The agent identifies the specific clients that are interacting with the bridge.
list_ai_clientsInspect specific threads
If you need to see the actual conversation history of a session, the AI pulls the specific message threads.
get_ai_session_threads
Try it
Copy these to start.
Use these prompts to begin your audit.
Starting points
These are opening lines. You should swap in specific session IDs or client names to get more targeted results.
Coder Connector
You're all set. Choose your MCP client and follow the setup instructions.
https://edge.vinkius.com/vk_preview_X1DE2rbvHJCSwP1lim38BmobcfpCklD913nroJvx/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 — Coder capabilities are ready to use.
{
"mcpServers": {
"coder-remote-dev-mcp": {
"url": "https://edge.vinkius.com/vk_preview_X1DE2rbvHJCSwP1lim38BmobcfpCklD913nroJvx/mcp"
}
}
}List all active AI sessions currently running in my dev environment.
Which AI models are being used most frequently right now?
Show me a list of all AI clients connected to the bridge.
Get the conversation threads for the session with ID 4592.
Claude
ChatGPT
Cursor
VS Code
Windsurf
Claude Code
JetBrains
Cline
Start here
Connect Coder (Remote Dev) once, then ask.
One link connects your AI client to the MCP. Your login stays encrypted on our side, and the full walkthrough lives on the Connector page.
Connect Coder (Remote Dev) to your AIFAQ
How this task behaves.
- 01
Can the AI delete active sessions?
No. The AI can only list and read session data. You must manage session termination through your own dev tools.
- 02
Can the AI change which model is being used?
No. The AI can only report on which models are currently in use.
- 03
Does this work with Cursor?
Yes, as long as you are using an MCP-compatible client like Cursor, Windsurf, or VS Code.
- 04
Can the AI see my private local files?
No. The AI only has access to the data exposed by the Coder (Remote Dev) MCP tools.
- 05
Can the AI modify my code during an audit?
No. This task is for monitoring and auditing usage, not for code manipulation.
More questions about Coder (Remote Dev)? The Connector page answers them. See everything the Coder (Remote Dev) Connector can do
Connect Coder to Claude, Cursor, ChatGPT & more
