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

# Tips and Tricks

> Get the most out of your Twin plan with these best practices

Twin is an AI company builder — you can create complex automations in minutes using only natural language. The agents you build can feel like giving a smart person 24/7 access to all your tools, so they can get things done for you.

We're still in the early days, and your results will vary depending on how you use it. These best practices will help you get the most out of Twin.

<Card title="TL;DR" icon="bolt" color="#F4D03F">
  Be specific, build once, then run on schedule with API-first instructions.
</Card>

**The key principles:**

* State the problem clearly and brainstorm with the agent to find the right solution
* Give enough context and mention the tools you use
* Leverage Twin's built-in tools (research, sheets, slides, email, and more)
* During build mode, monitor your agent and course-correct in the chat if it goes off track

***

## 1. How Credits Work

Paid credits start at **USD 0.01 / EUR 0.01 per credit**, with volume discounts on larger tiers. Credits are then consumed by model decisions (input/output/cache tokens) and paid tool usage when applicable.

<CardGroup cols={2}>
  <Card title="First runs cost more" icon="arrow-up" color="#F4D03F">
    Twin is discovering APIs, handling auth, and shaping reusable tools for the first time. This exploration is credit-intensive.
  </Card>

  <Card title="Later runs cost less" icon="arrow-down" color="#F4D03F">
    The agent reuses what it already learned — established tools, cached decisions, and proven paths.
  </Card>
</CardGroup>

<Note>Every decision counts, including retries. If an agent retries a failed step, that still consumes credits.</Note>

***

## 2. Why Scheduled Runs Are Cheaper

Build-style runs discover and design workflows. Run-style executions reuse established tools and instructions - making recurring runs **typically 3-10x cheaper** than builds.

<Steps>
  <Step title="Build and refine once" icon="hammer" color="#F4D03F">
    Get the agent working correctly through the chat editor. Iterate until the output is reliable.
  </Step>

  <Step title="Schedule it" icon="calendar" color="#F4D03F">
    Deploy the agent on a recurring schedule. Each scheduled run uses the lightweight run mode.
  </Step>

  <Step title="Improve with small updates" icon="pen" color="#F4D03F">
    When something needs to change, make small instruction updates instead of rebuilding from scratch. This preserves the agent's memory and keeps costs low.
  </Step>
</Steps>

***

## 3. API-First Beats Browser-First

Not all execution paths cost the same. Use this priority order whenever possible:

| Priority | Method                 | Why                                |
| -------- | ---------------------- | ---------------------------------- |
| 1        | **Built-in tools**     | Fastest, cheapest, most reliable   |
| 2        | **APIs**               | Deterministic and easy to maintain |
| 3        | **Scrapers**           | Good for static pages              |
| 4        | **Browser automation** | Last resort                        |

<CardGroup cols={2}>
  <Card title="APIs" icon="check" color="#F4D03F">
    Faster, more deterministic, and easier to maintain. Always preferred when available.
  </Card>

  <Card title="Browser automation" icon="xmark" color="#F4D03F">
    Slower, more brittle, and more expensive. Use only when there's no API or scraper alternative.
  </Card>
</CardGroup>

<Tip>Mention the tools and APIs you already use in your instructions. This helps Twin choose the most efficient path.</Tip>

***

## 4. Better Instructions = Lower Costs + Better Quality

The quality of your instructions directly impacts both cost and accuracy. Always include:

| Detail                    | What to specify                                              |
| ------------------------- | ------------------------------------------------------------ |
| **Source app(s)**         | Exactly where data comes from                                |
| **Filters & constraints** | Date range, owners, statuses, thresholds                     |
| **Output format**         | Table, CSV, JSON, email summary, doc, slide deck             |
| **Destination**           | Where the final output should go (Sheets, email, docs, etc.) |
| **Definition of done**    | One sentence that says what success looks like               |

### Prompt template

<Card title="Example prompt" icon="file-lines" color="#F4D03F">
  Every weekday at 8:00 AM, pull yesterday's new leads from HubSpot, exclude duplicates by email, append to my "Inbound Leads" Google Sheet, then send me a 5-bullet summary email with total count and top 3 companies.

  **Done when** the sheet is updated and the email is sent.
</Card>

***

## 5. Brainstorm and Debug to Keep Agents Efficient

Don't just build and forget — actively improve your agents over time.

<CardGroup cols={2}>
  <Card title="Brainstorm tasks" icon="lightbulb" color="#F4D03F">
    Use Twin to discover high-ROI workflows before building. Ask it what's possible with the tools you already use.
  </Card>

  <Card title="Debug last run" icon="bug" color="#F4D03F">
    When costs feel high or results are weak, ask Twin to analyze what happened and suggest improvements.
  </Card>
</CardGroup>

Ask Twin explicitly:

* "Where did this run spend most credits?"
* "What should I change to reduce cost by 30%?"

Review cost by run and by agent in the [credits dashboard](https://builder.twin.so/credits), then iterate your goal and instructions.

***

## 6. Keep Agents Focused to Avoid Context Bloat

Long-running agents accumulate context with every message and decision. The more steps an agent takes in a single session, the larger its context window grows — and the more tokens (credits) each subsequent step consumes.

<Warning>An agent that runs 50+ steps in a single build can cost significantly more per step toward the end than at the beginning, because it carries the full conversation history.</Warning>

To keep costs predictable:

* **Break complex workflows into multiple focused agents** inside a [Workspace](/workspaces) — the [Orchestrator](/orchestrator) can help you split them
* **Avoid open-ended goals** like "research everything about X" — set clear boundaries and a definition of done
* **Use Run mode** for recurring tasks — it starts fresh with the agent's memory, avoiding context buildup

***

## Built-In Tools

These are available out of the box and unlock a wide range of workflows without any setup:

<div className="tools-grid">
  <CardGroup cols={3}>
    <Card title="web_search" icon="magnifying-glass" color="#F4D03F">
      Fetch current info, API docs, and web answers
    </Card>

    <Card title="scrape" icon="file-code" color="#F4D03F">
      Pull clean content from specific pages
    </Card>

    <Card title="deep_research" icon="book-open-reader" color="#F4D03F">
      Multi-source research with async polling
    </Card>

    <Card title="stock_market" icon="chart-line" color="#F4D03F">
      Quotes, historical prices, and market news
    </Card>

    <Card title="send_email" icon="envelope" color="#F4D03F">
      Alerts, summaries, and outbound updates
    </Card>

    <Card title="build_chart" icon="chart-bar" color="#F4D03F">
      Generate chart images for reports
    </Card>

    <Card title="scrape_platforms" icon="hashtag" color="#F4D03F">
      Scrape social, job, map, and marketplace platforms
    </Card>

    <Card title="sheets" icon="table" color="#F4D03F">
      Create, read, update, and export Google Sheets
    </Card>

    <Card title="slides" icon="presentation-screen" color="#F4D03F">
      Generate presentation decks (PDF output)
    </Card>

    <Card title="docs" icon="file-lines" color="#F4D03F">
      Create and edit formatted Google Docs
    </Card>

    <Card title="generate_media" icon="image" color="#F4D03F">
      Create AI images and videos for campaigns and content
    </Card>
  </CardGroup>
</div>

<Tip>You don't need to tell Twin which tool to use — just describe what you need, and it will pick the right one. But mentioning a specific tool can help if you have a preference.</Tip>

***

## Quick Recap

<CardGroup cols={3}>
  <Card title="Be specific" icon="crosshairs" color="#F4D03F">
    Clear goals with filters, formats, and destinations
  </Card>

  <Card title="Build once" icon="hammer" color="#F4D03F">
    Refine until reliable, then schedule
  </Card>

  <Card title="API-first" icon="code" color="#F4D03F">
    Prefer APIs and built-in tools over browser automation
  </Card>
</CardGroup>

<Card title="Start Building" icon="arrow-right" href="https://builder.twin.so/" color="#F4D03F">
  Apply these tips in the Twin App
</Card>