Agentverse Manage

Overview

Manage your hosted agents on Agentverse. List all agents, start/stop them, read their logs, and delete agents you no longer need.

When to Use

• User asks "list my agents" or "what agents do I have"
• User asks to "start/stop/restart" an agent
• User asks to "show agent logs" or "what is my agent doing"
• User asks to "delete" or "remove" an agent
• You need to check agent status after deploying

Prerequisites

• AGENTVERSE_API_KEY environment variable set
• Python 3.8+ with requests

Quick Steps

List all agents

bash
python3 scripts/manage_agents.py list

Start an agent

bash
python3 scripts/manage_agents.py start --agent agent1q...

Stop an agent

bash
python3 scripts/manage_agents.py stop --agent agent1q...

View logs

bash
python3 scripts/manage_agents.py logs --agent agent1q...

Delete an agent

bash
python3 scripts/manage_agents.py delete --agent agent1q...

Output Examples

List

json
{
  "status": "success",
  "total": 3,
  "agents": [
    {
      "name": "my-image-relay",
      "address": "agent1q...",
      "running": true,
      "compiled": true,
      "domain": "",
      "wallet_address": "fetch1...",
      "created": "2026-04-20T10:00:00Z",
      "updated": "2026-04-20T15:30:00Z"
    }
  ]
}

Logs

json
{
  "status": "success",
  "agent": "agent1q...",
  "total_entries": 15,
  "showing": 3,
  "logs": [
    {"timestamp": "2026-04-20T15:30:00Z", "message": "Agent started", "level": "info"},
    {"timestamp": "2026-04-20T15:30:01Z", "message": "Sending message...", "level": "info"},
    {"timestamp": "2026-04-20T15:30:35Z", "message": "RESULT:{...}", "level": "info"}
  ]
}

API Reference

| Action | Method | Endpoint | |--------|--------|----------| | List agents | GET | /v1/hosting/agents | | Get agent | GET | /v1/hosting/agents/{address} | | Start | POST | /v1/hosting/agents/{address}/start | | Stop | POST | /v1/hosting/agents/{address}/stop | | Delete | DELETE | /v1/hosting/agents/{address} | | Get logs | GET | /v1/hosting/agents/{address}/logs/latest |

Edge Cases

• Agent already running: Starting a running agent is safe (no-op or restarts)
• Agent already stopped: Stopping a stopped agent is safe (no-op)
• Delete running agent: Stop it first, then delete
• Empty logs: Agent hasn't produced output yet, or was just started