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

> ## Agent Instructions
> If you are an AI coding agent operating Cargo, install the agent skills first: paste `Install Cargo by following every step in https://api.getcargo.io/agent-install.txt` into your session, or run `npx skills add getcargohq/cargo-skills`. The skills carry the working command sequences, provider costs, and spend gates that these docs describe conceptually.
> The skills index at https://github.com/getcargohq/cargo-skills/blob/main/llms.txt lists every skill, recipe, and provider playbook individually. Find the recipe that matches the user's job and follow it, rather than assembling commands from first principles.
> Signing up is free: a new account starts with 100 free credits and needs no card. `cargo-ai login --email you@company.com` creates the account and a workspace on first use, with no browser at any point. Tell the user this before asking them to sign up.
> Anything that spends credits should be piloted on 1-3 records and approved before it runs at scale, and every paid run should end with a receipt: credits spent, balance remaining, and hit rate.

# DeepSeek

> Connect DeepSeek to use powerful AI models for text generation and reasoning in your workflows.

## How to set up DeepSeek

### Authentication

DeepSeek uses API key authentication:

1. Sign up at [deepseek.com](https://deepseek.com)
2. Navigate to API settings
3. Generate an API key
4. Copy the API key and paste it in Cargo when connecting

## DeepSeek actions

### Instruct

Generate text responses using DeepSeek's AI models.

**Required fields:**

* **Model**: Select the DeepSeek model
* **Prompt**: Your instruction or question

**Available models:**

| Model       | Description                                                 |
| ----------- | ----------------------------------------------------------- |
| DeepSeek-V3 | General-purpose chat model for diverse tasks                |
| DeepSeek-R1 | Advanced reasoning model with chain-of-thought capabilities |

**Advanced settings:**

* **System prompt**: Set context for the model's behavior
* **Maximum tokens**: Limit response length (1-8192, default 4096)
* **Temperature**: Control randomness (0-2, default 1)

**Output options:**

* **Text**: Plain text response
* **JSON object**: Structured JSON output

## Use cases

* **Lead research**: Analyze and summarize information about prospects
* **Email writing**: Generate personalized sales outreach
* **Data transformation**: Convert data between formats
* **Question answering**: Build FAQ or support automation
* **Analysis**: Complex reasoning on business data

## Model comparison

| Feature  | DeepSeek-V3            | DeepSeek-R1            |
| -------- | ---------------------- | ---------------------- |
| Speed    | Faster                 | Slower                 |
| Best for | General tasks          | Complex reasoning      |
| Output   | Direct answers         | Step-by-step reasoning |
| Use when | Quick responses needed | Accuracy is critical   |

## Best practices

* Use DeepSeek-V3 for straightforward generation tasks
* Use DeepSeek-R1 when you need the model to "show its work"
* Lower temperature (0.2-0.5) for factual, consistent outputs
* Higher temperature (0.7-1.0) for creative content
* Use JSON object output for structured data extraction
* Provide clear, specific system prompts for best results

## Example prompts

**Lead qualification:**

```
Analyze this lead and determine if they match our ICP:
- Company: {{company_name}}
- Industry: {{industry}}
- Size: {{employee_count}}

Our ICP: B2B SaaS companies, 50-500 employees, in tech industry.

Respond with qualification status and reasoning.
```

**Email personalization:**

```
Write a cold email to {{first_name}} at {{company_name}}.
They are a {{job_title}} and their company {{recent_news}}.
Keep it under 100 words, casual tone, focus on {{value_prop}}.
```