text
stringlengths
0
59.1k
- Why did the agent respond this way?
- Which step failed?
- Where did latency or cost spike?
<video controls loop muted playsInline style={{width: '100%', height: 'auto'}}>
<source src="https://cdn.voltagent.dev/docs/voltop-docs/voltops-observability.mp4" type="video/mp4" />
Your browser does not support the video tag.
</video>
<br/>
<br/>
When you select a trace from the list, you can open the [Waterfall view](https://voltagent.dev/observability-docs/tracing/waterfall/) or the [Node-Based view](https://voltagent.dev/observability-docs/tracing/node-based/) and then dig into [Logs](https://voltagent.dev/observability-docs/tracing/logs/) and [Feedback](htt...
## Trace Filters
Filters help you find the right trace fast.
:::tip
For example use:
- Status + Duration to see where a failure started,
- Token usage + Cost to catch expensive runs, and
- User ID + Conversation ID to inspect a specific user flow.
:::
![Trace filters overview](https://cdn.voltagent.dev/docs/observability/tracing/overview-trace-filters.gif)
<br/>
- **Status**: isolate failed or retry-heavy runs.
- **Agent ID / Entity type**: narrow down which agent or step is involved.
- **Token usage / Cost**: find expensive or abnormal runs.
- **Duration**: spot slow traces and latency outliers.
- **Feedback source / key**: review feedback by source.
- **User ID / Conversation ID**: drill into one user journey.
<|endoftext|>
# source: VoltAgent__voltagent/website/actions-triggers-docs/overview.md type: docs
---
title: Actions & Triggers
slug: /
---
# Actions & Triggers
VoltOps Actions & Triggers connect your agents and workflows to external services:
- **Triggers** determine when your agents start running - responding to messages, emails, database changes, or cron schedules
- **Actions** push agent output to external systems - sending messages, creating records, or calling APIs
Both share the same credential system in the VoltOps console, so you authenticate once and use everywhere.
## How It Works
```mermaid
%%{init: {'theme':'dark', 'themeVariables': { 'primaryColor':'#10b981', 'primaryTextColor':'#fff', 'primaryBorderColor':'#10b981', 'lineColor':'#10b981', 'secondaryColor':'#1a1b1e', 'tertiaryColor':'#0f1011', 'background':'#0f1011', 'mainBkg':'#0f1011', 'nodeBorder':'#10b981', 'edgeLabelBackground':'#1a1b1e', 'labelCol...
flowchart LR
A["External Event<br/><small>Slack, Gmail, GitHub, etc.</small>"]
B["Trigger<br/><small>Starts your agent</small>"]
C["Agent/Workflow<br/><small>Processes the event</small>"]
D["Action<br/><small>Sends output to SaaS</small>"]
A -.detects.-> B
B -.executes.-> C
C -.calls.-> D
style A fill:#0f1011,stroke:#10b981,stroke-width:3px,color:#fff,rx:10,ry:10
style B fill:#0f1011,stroke:#10b981,stroke-width:3px,color:#fff,rx:10,ry:10
style C fill:#0f1011,stroke:#10b981,stroke-width:3px,color:#10b981,rx:10,ry:10
style D fill:#0f1011,stroke:#10b981,stroke-width:3px,color:#fff,rx:10,ry:10
linkStyle 0 stroke:#10b981,stroke-width:2px,color:#10b981
linkStyle 1 stroke:#10b981,stroke-width:2px,color:#10b981
linkStyle 2 stroke:#10b981,stroke-width:2px,color:#10b981
```
## Triggers
Triggers monitor external services and execute your agents when events occur.
| Provider | Description | Delivery Method |
| ------------------- | ------------------------------------------------------ | --------------- |
| **Slack** | Monitor workspace messages and channels | Webhook |
| **Gmail** | Watch for emails with specific labels | Polling |
| **Google Calendar** | React to calendar events (created, updated, cancelled) | Polling |
| **Google Drive** | Detect file or folder changes | Polling |
| **Airtable** | Detect record changes in bases and tables | Polling |
| **GitHub** | Respond to repository events (PRs, issues, commits) | Webhook |
| **Schedule** | Execute on cron expressions | Time-based |
See the [Triggers documentation](./triggers/overview.md) to get started.
## Actions
Actions send data from your agents to external systems with managed credentials.
| Provider | Description |