> ## Documentation Index
> Fetch the complete documentation index at: https://docs.collabase.ch/llms.txt
> Use this file to discover all available pages before exploring further.

# AI (local)

> Run prompts through any OpenAI-compatible language model including local Ollama instances.

# AI (local)

Run LLM prompts directly inside automations using your Collabase AI configuration or any OpenAI-compatible endpoint — including local Ollama instances with no API key required.

## Authentication

**Auth type:** API Key (optional for local Ollama)

Configure AI settings in **Admin Settings → AI Configuration**. The connector reads your configured base URL and API key automatically.

| Credential   | Description                                                                    |
| ------------ | ------------------------------------------------------------------------------ |
| **Base URL** | `http://localhost:11434/v1` for Ollama, `https://api.openai.com/v1` for OpenAI |
| **API Key**  | Leave empty for local Ollama. Required for OpenAI and other hosted providers.  |

## Actions

| Action             | Key inputs                                                        | Key outputs                |
| ------------------ | ----------------------------------------------------------------- | -------------------------- |
| **Generate Text**  | `model`, `systemPrompt`, `userPrompt`, `temperature`, `maxTokens` | content, model, tokensUsed |
| **Summarize Text** | `text`, `model`, `maxSentences`                                   | summary                    |

## Example

```
[Trigger: Test Case Failed]
    ↓
[AI: Summarize Text]
    text: {{testCaseTitle}} — {{testCaseDescription}}
    maxSentences: 2
    ↓
[Slack: Send Message]
    channel: C012QA
    text: "Test failure: {{previous.summary}}"
```

<Note>
  The **AI (local)** connector uses Collabase's system-wide AI configuration. To use a per-automation API key instead, use the **OpenAI** connector with explicit credentials.
</Note>