- COMPLIANCE
- SECURITY
- CONFIG
- HUMAN
Short answer
How can I audit my AI agent's activity?
Your AI pulls activity logs and agent configurations to identify unauthorized actions or policy violations. You get a clear view of what your agents are doing and how they are set up.
Audit outcomes
Where your audit data lands.
The AI processes raw data into these specific results.
COMPLIANCE
Policy verification
The AI compares agent actions against your defined rules. You get a report on whether the agent is following protocol.
SECURITY
Activity monitoring
The AI scans logs for suspicious patterns or unexpected tool usage. You see exactly when and how an agent acted.
CONFIG
Setup inspection
The AI retrieves current agent settings. You can check if permissions or instructions have changed unexpectedly.
HUMAN
Manual review
The AI flags specific log entries that require your attention. You decide if an action needs to be revoked or corrected.
The workflow
What your AI does when the request arrives.
The AI uses Runlayer tools to gather the data you need for a security review.
Gather logs
The AI pulls the recent history of actions taken by your agents.
get_audit_logsCheck settings
The AI looks up the specific configuration and instructions for a target agent.
get_agentReview rules
The AI lists the active security policies currently governing your environment.
list_policiesFlag issues
The AI compares the logs and settings against your policies to find mismatches.
get_audit_logslist_policies
Try it
Copy these to start.
Use these prompts to begin your audit.
Starting points
Replace the bracketed text with your specific 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"
}
}
}Show me the last 50 activity logs for the 'CustomerSupport' agent.
Check if the 'DataProcessor' agent is following the 'StrictDataPolicy'.
List the current configuration for the 'MarketingBot' agent.
Find any recent logs where an agent attempted to access restricted tools.
Claude
ChatGPT
Cursor
VS Code
Windsurf
Claude Code
JetBrains
Cline
Start here
Connect Runlayer once, then ask.
One connection gives your AI access to all your agent data. Your credentials stay encrypted and secure.
Connect Runlayer to your AIFAQ
How auditing works
- 01
Can the AI change my agent's settings?
No. The AI can only read configurations and logs. You must manually change settings in your dashboard.
- 02
Can the AI delete audit logs?
No. The tools only allow for retrieving and listing data, not deleting it.
- 03
How often does the AI see new logs?
The AI sees whatever data is available through the logs at the moment you ask the question.
- 04
Can I audit multiple agents at once?
Yes, you can ask the AI to compare logs across different agents or check multiple configurations in one session.
- 05
Does the AI know which policies are active?
Yes, it can list all active policies to use as a reference for its audit.
More questions about Runlayer? The Connector page answers them. See everything the Runlayer Connector can do
Connect Runlayer to Claude, Cursor, ChatGPT & more
