Documentation Index
Fetch the complete documentation index at: https://docs.shinzo.ai/llms.txt
Use this file to discover all available pages before exploring further.
The Shinzo Platform allows you to deploy and manage autonomous AI agents that can execute complex tasks. Agents run in isolated Kubernetes environments with persistent storage and can interact with external services via MCP servers.
Features
- Isolated Environments: Each agent runs in its own container with dedicated resources and persistent storage
- Messaging: Send messages to agents and receive responses via the API or connected channels
- Schedules & Triggers: Automate agent tasks with recurring cron schedules and event-driven webhook triggers
- Channel Integrations: Connect agents to Discord, Slack, and Telegram. WhatsApp support coming soon
- Filesystem API: Read, write, and manage files in the agent’s workspace
- MCP Support: Connect agents to MCP servers for external service integration
- Webhooks: Receive notifications when messages complete, fail, or when errors occur
Authentication
All agent endpoints accept either:
- JWT Token:
Authorization: Bearer <jwt_token>
- Platform API Key:
Authorization: Bearer <api_key>
Feature Flag
AI Agents must be enabled for your account. If you receive a 403 Forbidden error, contact support to enable this feature.
Endpoints
Agent Management
| Method | Endpoint | Description |
|---|
POST | /v1/agent/create | Deploy a new AI agent |
GET | /v1/agent/list | List all your agents |
GET | /v1/agent/{id} | Get agent details and configuration |
GET | /v1/agent/{id}/summary | Get agent summary (status, tokens, metadata) |
GET | /v1/agent/{id}/ready | Check if agent is ready to receive messages |
POST | /v1/agent/update/{id} | Update agent configuration |
POST | /v1/agent/delete/{id} | Delete an agent |
POST | /v1/agent/{id}/pause | Pause an agent |
POST | /v1/agent/{id}/resume | Resume a paused agent |
POST | /v1/agent/{id}/stop | Stop an agent (shuts down pod) |
POST | /v1/agent/{id}/status | Set agent status directly |
POST | /v1/agent/{id}/balance | Update agent token balance |
Messaging
| Method | Endpoint | Description |
|---|
POST | /v1/agent/{id}/messages | Send a message to an agent |
GET | /v1/agent/{id}/messages | List agent messages |
GET | /v1/agent/{id}/conversations | List conversations grouped by counterparty |
Channel Integrations
| Method | Endpoint | Description |
|---|
GET | /v1/agent/{id}/channels | List active channel bindings |
POST | /v1/agent/{id}/discord/link | Generate Discord link code |
GET | /v1/agent/{id}/discord/status | Check Discord link status |
DELETE | /v1/agent/{id}/discord/unlink | Unlink Discord account |
POST | /v1/agent/{id}/slack/link | Generate Slack link code |
GET | /v1/agent/{id}/slack/status | Check Slack link status |
DELETE | /v1/agent/{id}/slack/unlink | Unlink Slack account |
POST | /v1/agent/{id}/telegram/link | Generate Telegram link code |
GET | /v1/agent/{id}/telegram/status | Check Telegram link status |
DELETE | /v1/agent/{id}/telegram/unlink | Unlink Telegram account |
Schedules & Triggers
| Method | Endpoint | Description |
|---|
POST | /v1/agent/{id}/schedules | Create a cron schedule |
GET | /v1/agent/{id}/schedules | List all schedules |
PATCH | /v1/agent/{id}/schedules/{scheduleId} | Update a schedule |
POST | /v1/agent/{id}/schedules/{scheduleId}/toggle | Toggle schedule enabled state |
DELETE | /v1/agent/{id}/schedules/{scheduleId} | Delete a schedule |
GET | /v1/agent/{id}/schedules/{scheduleId}/executions | Get schedule execution history |
POST | /v1/agent/{id}/triggers | Create a webhook trigger |
GET | /v1/agent/{id}/triggers | List all triggers |
PATCH | /v1/agent/{id}/triggers/{triggerId} | Update a trigger |
DELETE | /v1/agent/{id}/triggers/{triggerId} | Delete a trigger |
POST | /v1/webhooks/triggers/{triggerId} | Inbound webhook endpoint (no auth header) |
See Schedules & Triggers Overview for detailed documentation and usage examples.
Filesystem
See Filesystem Overview for complete filesystem API documentation.
Agent States
| State | Description |
|---|
active | Agent is running and accepting messages |
paused | Agent is paused and will not process new messages |
stopped | Agent has been stopped |
error | Agent encountered an error |