# Automations (Beta)
Source: https://docs.happycapy.ai/en/features/automations

Run recurring work in a Project on a schedule

<Note>
  **Beta:** Automations is in early access. Functionality will continue to improve.
</Note>

## What is an Automation?

An Automation is a scheduled prompt that runs inside a Project. It can use the Project's files, session context, and instructions, then save or send the result.

Automations are useful when the work is repeatable, has a clear schedule, and produces an output you can review.

## When to use Automations

Use Automations for recurring work with clear inputs and a repeatable output:

* Daily or weekly summaries
* Project status reports
* Scheduled file processing
* Recurring research or monitoring
* Periodic GitHub, Notion, or email checks
* Work that should continue while you are away

<Note>
  Automations work best when the prompt explains the goal, source material, output format, and what to do if there is nothing new.
</Note>

## Create an Automation

You can create an Automation from the sidebar. You can also ask Happycapy to create one from a Project conversation.

<div>
  <div>
    <span>Automations</span>
    <span>Schedule recurring work in a Project.</span>
  </div>

  <img alt="Create Automation dialog" />
</div>

To use the sidebar, click **Automations**, then **Create Automation**.

In a Project conversation, describe what should run and when:

```text theme={null}
Create an Automation that summarizes this Project every Friday at 4 PM and saves the report to outputs/.
```

Fill in:

* **Name** — a short label, such as "Morning summary"
* **Project** — the Project where the Automation should run
* **Session binding** — the main session or a session inside that Project
* **Schedule** — daily at a specific time, or on an interval
* **Prompt** — the instruction Happycapy will run
* **Run immediately** — run once now in addition to the schedule

Click **Create** to save it.

## What each field means

* **Project** keeps the Automation tied to the right workspace and files.
* **Session binding** decides which session context the prompt should continue from.
* **Schedule** controls when the Automation runs.
* **Prompt** is the instruction Happycapy executes each time.
* **Run immediately** lets you test the prompt once before relying on the schedule.

## Write a reliable prompt

Automations run without you typing the prompt again, so the prompt should be self-contained.

* State the goal and output format.
* Mention the source files, folders, or services to check.
* Tell Happycapy where to save or send the result.
* Include fallback behavior, such as "If there is nothing new, write a short no-update note."

## Examples

### Morning brief

```text theme={null}
Fetch trending news in AI, startups, and product design.
Filter out clickbait. Summarize each item in one sentence.
Generate an HTML brief and send it to my email.
```

### Project status report

```text theme={null}
Summarize all work done in this Project over the past week.
List completed tasks, open issues, and next steps.
Generate a report and save it to outputs/.
```

### GitHub review check

```text theme={null}
Check all open PRs in my-org/my-repo.
List any that need my review or have new comments.
Post a summary to this Project.
```

### File processing

```text theme={null}
Every weekday, check the uploads folder for new CSV files.
Clean and summarize any new data.
Save the summary to outputs/daily-data-summary.md.
If there are no new files, write a short no-update note.
```

<Tip>
  The prompt runs like a normal message in a session. Anything you can do manually in a Project can usually become an Automation when the steps are repeatable.
</Tip>


# Capy Mail
Source: https://docs.happycapy.ai/en/features/capy-mail

Bring email into your Happycapy workflow.

## What is Capy Mail?

Capy Mail gives your Happycapy account an email address that can send and receive email.

Use Capy Mail when you want Happycapy to send email from a Project, send the same update to multiple recipients, receive replies, or keep an email thread inside the same Happycapy conversation.

You can also start a new Happycapy agent conversation by email. After the conversation is created, replies in the same email thread continue in that conversation.

Capy Mail is not a separate email app. It makes email part of the work Happycapy is already doing.

## When to use Capy Mail

* Send email from a Project
* Send the same update, report, or notice to multiple recipients
* Start a new Happycapy conversation by email
* Receive replies and continue with the same context
* Ask Happycapy to summarize, classify, answer, or follow up on email work
* Pair email with Automations for recurring reports or follow-ups

## Set up Capy Mail

Go to **Settings → Capy Mail**. Your Capy Mail address is shown there, using the `capymail.ai` domain.

<div>
  <div>
    <span>Capy Mail</span>
    <span>Send work to Happycapy by email. Get the result back when the cloud task finishes.</span>
  </div>

  <div>
    <div>
      <div>
        <span />

        <span />

        <span />

        <span>happycapy.ai</span>
      </div>

      <div>
        <img alt="Happycapy completes work and sends email" />
      </div>
    </div>

    <div>
      <div>
        <div>
          <span>✓</span>
          <span>Your @capymail.ai address receives the task</span>
        </div>

        <div>
          <span>✓</span>
          <span>Happycapy completes the task in the cloud</span>
        </div>

        <div>
          <span>✓</span>
          <span>Results and attachments are emailed back</span>
        </div>
      </div>
    </div>

    <div>
      <div />

      <img alt="A report received by email on mobile" />
    </div>
  </div>
</div>

Add the email addresses you trust to **Allowed Senders**. Capy Mail only processes email from those addresses.

To add an allowed sender:

1. Open **Settings → Capy Mail**
2. Type an email address into the **Add email address** field
3. Click the **+** button

You can remove a sender at any time.

## Ask Happycapy to send email

You can ask Happycapy to send email from a Project. It can send to one recipient or to multiple recipients.

Happycapy can handle the email workflow: drafting the message, using Project context, attaching or referencing Project outputs, and sending the final email.

## Start a conversation by email

Email can start a new Happycapy agent conversation. After the conversation is created, replies in the same email thread continue in that conversation.

This is useful for follow-ups, feedback, and lightweight support workflows. Happycapy can summarize incoming email, prepare replies, or continue answering with the context from that thread.

## Use with Automations

When email should be sent on a schedule or become part of a recurring workflow, pair Capy Mail with [Automations](/en/features/automations).

For example, ask Happycapy to send recurring reports, follow up with a group of contacts, or handle incoming email for a Project.


# CLAUDE.md
Source: https://docs.happycapy.ai/en/features/claude-md

Use CLAUDE.md to give Happycapy stable instructions for a Project.

## What is CLAUDE.md?

`CLAUDE.md` is a project instruction file. Use it to tell Happycapy what to remember, how to work, and what rules to follow when it returns to the same Project.

You do not need to treat it like code. Think of it as the standing brief for a Project: the audience, tone, output rules, file locations, and review habits that should stay true across sessions.

Happycapy can create or update `CLAUDE.md` for you. You can also open **Files**, select the file, and edit it directly.

After you create a Project and start working, open **Files** to see the Project workspace. `CLAUDE.md` sits alongside uploaded source files and generated outputs, so future sessions can reuse the same instructions.

<div>
  <div>
    <div>
      <span>
        <span>
          <svg>
            <path />
          </svg>
        </span>

        Files
      </span>
    </div>

    <div>
      <div>Project workspace</div>

      <div>
        <div><span>›</span><span>memory/</span></div>
        <div><span>›</span><span>outputs/</span></div>
        <div><span>›</span><span>uploads/</span></div>

        <div>
          <span>
            <svg>
              <path />

              <path />

              <path />

              <path />

              <path />
            </svg>
          </span>

          <span>CLAUDE.md</span>
        </div>
      </div>
    </div>
  </div>

  <div>
    <div>
      <span>CLAUDE.md</span>
      <span>Editable</span>
    </div>

    <div>
      <div>Project instructions</div>
      <p>Tell Happycapy what to remember and follow when it works in this Project.</p>
      <div>Add stable project rules below this line.</div>
    </div>
  </div>
</div>

## When to use it

Add something to `CLAUDE.md` when it should help future sessions in the same Project.

Good reasons include:

* You corrected Happycapy for the same thing more than once.
* The Project has rules you do not want to repeat every time.
* Future work needs the same audience, tone, file location, or review standard.
* Someone else opening the Project would need this context to get useful results.

Keep one-off task details in the current message. Put stable Project context in `CLAUDE.md`.

## Write instructions Happycapy can follow

`CLAUDE.md` guides Happycapy. It is not a hard safety switch, so clear and specific instructions work better than broad preferences.

| Vague                           | Better                                                                         |
| ------------------------------- | ------------------------------------------------------------------------------ |
| Write better copy.              | Write concise customer-facing copy in plain English.                           |
| Save the file somewhere useful. | Save final files in `/outputs`.                                                |
| Check your work.                | Before finishing, list changed files and remaining risks.                      |
| Follow the brand.               | Use the uploaded brand guide as the source of truth for voice and terminology. |

A useful `CLAUDE.md` is usually short, grouped with headings, and easy to scan.

```text theme={null}
Add this to this Project's CLAUDE.md:

## Writing
- Write customer-facing copy in clear, concise English.
- Avoid hype unless it is tied to a concrete user task.

## Files
- Save final deliverables in /outputs.
- Include the date in filenames for reports and exports.

## Review
- Ask before changing pricing, legal, or public launch language.
- Before finishing, list changed files and any remaining risks.
```

You can also ask Happycapy to review the file:

```text theme={null}
Review this Project's CLAUDE.md. Tell me what instructions are unclear, stale, or too broad.
```

## Project and global instructions

Most Projects only need one `CLAUDE.md` file in the Project workspace. Use broader locations only when the distinction matters.

| Location                       | Use it for                                                                                                                 | Keep out                                                 |
| ------------------------------ | -------------------------------------------------------------------------------------------------------------------------- | -------------------------------------------------------- |
| `CLAUDE.md` in a Project       | Instructions for one Project                                                                                               | Personal defaults that should affect every Project       |
| `/home/node/.claude/CLAUDE.md` | Personal defaults you want Happycapy to reuse across Projects, such as language, tone, summary style, and progress updates | Project facts, customer data, and temporary task details |

<Tip>
  Want the same preference to apply across Projects? Ask Happycapy to update `/home/node/.claude/CLAUDE.md`. For example: "Set my global CLAUDE.md so Happycapy defaults to Chinese, keeps progress updates concise, and asks before changing public-facing copy."
</Tip>

If you are unsure where something belongs, start with the Project `CLAUDE.md`. Keep the file focused on instructions that should apply to most future work in that Project.

## Related resources

* [Projects](/en/features/projects): organize sessions and files around ongoing work
* [Files](/en/features/files): manage uploaded and generated files
* [Skill Store](/en/features/skill-store): install reusable workflows
* [Automations](/en/features/automations): schedule recurring work


# Connect My Mac
Source: https://docs.happycapy.ai/en/features/connect-mac

Use your Mac when a task needs local files, apps, or tools.

## What is Connect My Mac?

Most tasks can run inside Happycapy's cloud sandbox.

Use **Connect My Mac** when a task needs files on your Mac, local apps, or tools you already have installed. Once connected, Happycapy can run tasks on your Mac and connect your cloud Project with your local environment.

You can connect more than one Mac and disconnect a device at any time.

## When to use Connect My Mac

* You need to work with files stored on your Mac
* You need to use an app installed on your Mac
* You need to run a local script, CLI, or development environment
* You want to move work between a cloud Project and your local setup

## Connect a Mac

Open **Settings**, then go to **Connect My Mac**.

<div>
  <div>
    <span>Connect My Mac</span>
    <span>Connect local Macs for files, apps, and tools.</span>
  </div>

  <img alt="Connect My Mac settings" />
</div>

To connect a device:

1. Click your avatar in the bottom-left corner
2. Open **Settings**
3. Go to **Connect My Mac**
4. Click **+ Connect a Mac**
5. Follow the setup instructions

Once connected, your Mac appears in the device list.

## What Happycapy can do on your Mac

With a connected Mac, Happycapy can:

* read, organize, and save local files
* run local commands, scripts, and CLI tools
* use desktop apps to complete tasks
* continue work between the cloud sandbox and your Mac

## Multiple devices

You can connect multiple Macs. Each device appears separately in Settings, so you can choose the right machine for the task.


# Dynamic Workflows
Source: https://docs.happycapy.ai/en/features/dynamic-workflows

Run complex Project tasks through staged execution, subagents, and saved outputs.

## What are Dynamic Workflows?

Dynamic Workflows help Happycapy handle complex Project work as a staged run instead of one long reply. Use them when a task needs planning, context gathering, subtasks, parallel subagents, review, and saved files.

You still start from a normal Project message. Enable Dynamic Workflows in Settings first, then describe the outcome and the final files you want. Happycapy can plan the run, split work into subtasks, use subagents when helpful, review intermediate results, and save the final artifacts back to the Project.

## How Dynamic Workflows run

<div>
  <div>
    <span>From request to outputs</span>
    <span>Plan, split, review, and save the final files.</span>
  </div>

  <img alt="Dynamic Workflows turn one Project request into planning, subtasks, review, and saved output files" />
</div>

## How to use Dynamic Workflows

### Enable it in Settings

Dynamic Workflows are an experimental capability. Open **Settings**, go to **Experimental**, and turn on **Dynamic workflows** before using them in a Project.

<div>
  <div>
    <span>Settings</span>
    <span>Open Experimental and turn on Dynamic workflows.</span>
  </div>

  <div>
    <img alt="Happycapy Settings Experimental page with Dynamic workflows enabled" />
  </div>
</div>

<Warning>
  Dynamic Workflows can use a lot of credits. They may run multiple stages, read more context, call more tools, create intermediate artifacts, and use subagents in parallel. Save them for complex, multi-step work where redoing the task would be expensive.
</Warning>

### Start from a Project conversation

Do not assign each subagent yourself. Tell Happycapy what work you need done and which files you want at the end. Happycapy decides how to plan, split, review, and save the work.

```text theme={null}
Use Dynamic Workflows for this task.

I am preparing to launch an AI video script tool for independent creators.
Please research and plan the launch from these angles:

- target users
- competitors
- pricing
- content channels
- first features
- launch risks

Create the following files:

- outputs/market-entry-brief.md
- outputs/competitor-comparison.csv
- outputs/two-week-launch-plan.xlsx
- outputs/landing-page-outline.html
- outputs/social-posts.json
```

### Watch the run

<div>
  <div>
    <span>Example run</span>
    <span>A run can show subtasks, active agents, checks, and files being created.</span>
  </div>

  <img alt="Happycapy running Dynamic Workflows with subtasks, active agents, checks, and output files" />
</div>

## What to know

### Use them for complex work

Use Dynamic Workflows for work that has a clear goal but needs more than a single answer:

* Market, product, or competitor research that needs synthesis
* Multi-step work that creates several files in different formats
* Reports, spreadsheets, outlines, pages, or JSON assets that need review before saving
* Tasks where later steps depend on what Happycapy discovers earlier
* Work where a rushed first draft would create a lot of rework

### Avoid them for small tasks

Avoid Dynamic Workflows for simple questions, short copy edits, quick translations, and small one-file changes. A normal Project message is cheaper and faster for those.

### Let Happycapy handle the split

During a run, Happycapy turns your request into a plan, splits work into subtasks, uses subagents when parallel work helps, checks intermediate results, and saves the final files back to the Project.

That is what makes the process dynamic: the route can adjust while the task is running, instead of following a fixed template.

### Start with a focused scope

Start with a focused scope before handing over a very broad task. Name the final files, review the first run, then expand the workflow when the pattern is stable.

If you repeat the same workflow often, ask Happycapy to turn the pattern into a reusable Project instruction or Skill. If it should run on a schedule, connect the stable prompt to an Automation.

## Related

* [Projects](/en/features/projects): keep related work, files, and instructions together
* [Automations](/en/features/automations): run recurring Project work on a schedule


# Editable Spreadsheets
Source: https://docs.happycapy.ai/en/features/editable-spreadsheets

Open CSV files as editable tables and work on them with Happycapy.

Editable Spreadsheets let you open CSV files as table views directly from the Project Explorer. You can inspect rows, search through structured data, edit cells directly, and keep working with Happycapy in the same conversation.

Use this when a spreadsheet needs both human review and agent help: clean rows, fix values, rename columns, add calculated fields, summarize data, or prepare a report from the edited file.

## Open a CSV from Explorer

Inside a Project, open **Explorer** and select a supported CSV file. Happycapy opens the file on the right as an interactive table instead of a plain text preview.

<div>
  <div>
    <div aria-label="Project file browser">
      <div>
        <span>
          <span />

          Explorer
        </span>
      </div>

      <div>
        <div>File Tree</div>

        <div aria-label="Project workspace files">
          <div>
            <span>⌄</span>
            <span>memory</span>
          </div>

          <div>
            <span>›</span>
            <span>outputs</span>
          </div>

          <div>
            <span>›</span>
            <span>uploads</span>
          </div>

          <div>
            <span />

            <span>yc\_companies.csv</span>
          </div>
        </div>
      </div>
    </div>

    <div>
      <div>
        <table aria-label="Editable CSV table preview">
          <colgroup>
            <col />

            <col />

            <col />

            <col />
          </colgroup>

          <thead>
            <tr>
              <th>Name</th>
              <th>Website</th>
              <th>One-liner</th>
              <th>Industry</th>
            </tr>
          </thead>

          <tbody>
            <tr>
              <td>Ablinitio Bio</td>
              <td>abinitio-bio.com</td>
              <td>Intelligence layer for drug manufacturing</td>
              <td>Healthcare</td>
            </tr>

            <tr>
              <td>Adialante</td>
              <td>adialante.com</td>

              <td>
                Cancer screening without barriers<span />
              </td>

              <td>Healthcare</td>
            </tr>

            <tr>
              <td>Akkari</td>
              <td>akkari.io</td>
              <td>Autonomous customer operations</td>
              <td>B2B</td>
            </tr>
          </tbody>
        </table>
      </div>
    </div>
  </div>
</div>

## What You Can Do

You can:

* click a cell and edit the value directly
* search across the table
* scroll through rows and columns
* review row and column counts
* keep the file open while asking Happycapy to analyze or transform it

## Edit by Hand

Open **Explorer** in the Project, select the CSV file, then edit the table on the right. Select a cell, change the value, and continue reviewing the data without leaving the Project conversation.

Manual edits are useful when you already know the correction:

* fix a company name
* normalize a category
* adjust a date or status
* correct a short description
* inspect a specific row before asking Happycapy to continue

## Ask Happycapy to Edit

You can also ask Happycapy to modify the table for you. Describe the transformation you want, and mention the file name when helpful.

```text theme={null}
Open yc_companies.csv, normalize the Industry column, merge similar categories,
and save the cleaned version as outputs/yc_companies_cleaned.csv.
```

Happycapy can help with:

* cleaning inconsistent values
* filtering rows
* renaming or adding columns
* extracting structured fields from text
* calculating new values
* merging or comparing CSV files
* summarizing the table into a report

## Combine Manual and Agent Edits

Editable Spreadsheets work best when you use both modes together.

1. Open the CSV from the Project Explorer and inspect the data.
2. Edit obvious cells yourself.
3. Ask Happycapy to clean or transform the rest.
4. Review the updated file.
5. Ask Happycapy to create a report, chart, or export from the final table.

<Tip>
  If the table is important, ask Happycapy to save a cleaned copy instead of overwriting the original file.
</Tip>

## Related

* [Files](/en/features/files): find uploads, output files, and Project files
* [Projects](/en/features/projects): keep related sessions and files together
* [Dynamic Workflows](/en/features/dynamic-workflows): handle larger multi-stage data tasks


# Files
Source: https://docs.happycapy.ai/en/features/files

Find uploads, output files, and project files in one place

## What are Files?

Files is the file browser for your Happycapy sandbox. Use it to find uploaded materials, Project files, and output files Happycapy creates while working.

Files is especially useful after tasks that produce assets, reports, code, data exports, images, or videos.

## Open Files

<div>
  <div>
    <span>Files</span>
    <span>Browse uploads, output files, and Project files.</span>
  </div>

  <img alt="Files browser" />
</div>

Open **Settings**, then click **Files**.

You can browse files from:

* **Home** — files in your sandbox
* **Projects** — files grouped by Project
* **Skills** — installed Skill files

## Find and manage files

Use search to find files by name. Switch between grid and list view when you need to scan images, folders, or file details.

You can open files, preview supported formats, rename, move, delete, or download files from the file browser.

## Upload files

Attach files from the input box with **+**, or drag files into the conversation. Happycapy can use those files as context for the task.

Happycapy can work with many common file types, including documents, spreadsheets, presentations, images, videos, archives, code files, Markdown, CSV, JSON, and more. If you are unsure, upload the file and ask Happycapy to read or inspect it.

## Work with outputs

When Happycapy creates files, they are saved in your sandbox and can be found in Files.

For example, you can ask:

```text theme={null}
Package the images in this Project into a zip file so I can download them.
```

## Files and Projects

Project files stay with the Project. Sessions inside the same Project can read and write the same files, so you can continue work across sessions without re-uploading the same files.

See [Projects](/en/features/projects) for how Projects organize sessions and files.

## Locate files from a Project or session link

When a conversation link matches either format below, the value after `/s/` is the Project ID. Use that Project ID when searching Files for JSONL logs or workspace files.

<div>
  <div>Conversation link</div>
  <div>https\://<a href="https://happycapy.ai">happycapy.ai</a>/s/<strong>d94c8b54-161d-4278-a3f5-16e5046d66cd</strong></div>

  <br />

  <div>Conversation link with a sub-session</div>
  <div>https\://<a href="https://happycapy.ai">happycapy.ai</a>/s/<strong>d94c8b54-161d-4278-a3f5-16e5046d66cd</strong>?sub=48c366cc-a0e0-4bf4-ac56-4fa52c7cd7a7</div>

  <br />

  <div>Project ID</div>
  <div><strong>d94c8b54-161d-4278-a3f5-16e5046d66cd</strong></div>
</div>

### Find JSONL logs

Open **Files**, go to `Home > .claude > projects`, and search for the **Project ID** from the link.

The matching folder contains `.jsonl` logs for conversations under that Project. Each `.jsonl` represents one conversation, including the main session or a sub-session.

### Find the workspace files

Open **Files** and use either route:

* Open **Projects** and search for the same **Project ID**.
* Or go to `Home > a0 > workspace` and search for the same **Project ID**.

The matching Project or workspace folder contains files Happycapy created or edited under that Project.

### If something went wrong

You can inspect those files yourself, or open a new session and ask Happycapy:

```text theme={null}
Under /home/node/.claude/projects/, find the folder for d94c8b54-161d-4278-a3f5-16e5046d66cd (Project ID), back up the corresponding JSONL log, diagnose what went wrong, fix it, and resume this conversation.
```


# Image Creation & Editing
Source: https://docs.happycapy.ai/en/features/image-generation

Create, edit, remix, and prepare reference images inside a Project

## What is Image Creation & Editing?

Image Creation & Editing lets you generate new images, edit existing images, and combine visual references inside a Project. Use it for covers, product visuals, campaign assets, reference frames, and other still images you want to keep with your files.

Start from the result you need, not from a model name. Happycapy can save the output to Files and keep prompts, references, and generated assets together for later sessions.

## Three ways to start

Choose the input you already have: text, one image, or multiple references.

### 1. Start from text

Describe the image you want. Happycapy can generate it and save it with the rest of your Project files.

<div>
  <div>
    <div>
      <div>Text to image</div>
      <h3>Create a docs cover from text</h3>
      <p>Start with the finished image you want. Happycapy can generate the artwork and save it with the rest of your Project files.</p>

      <div>
        <div>
          <p>Ask Happycapy</p>
          <code>Create a 4:5 Happycapy-style illustration for a docs page about image generation.Show a calm capybara arranging blank image cards into one finished artwork. Use bright daylight, clean editorial illustration, paper grain, and no readable text.Save it as outputs/image-generation-cover.png.</code>
        </div>

        <div>
          <img alt="Happycapy-style generated illustration for an image generation docs page" />
        </div>
      </div>
    </div>
  </div>
</div>

### 2. Edit one image

Upload an image and ask Happycapy to change details, layout, background, style, or local areas while keeping the parts you care about.

<div>
  <div>
    <div>
      <div>Edit an image</div>
      <h3>Turn a plain product photo into a campaign poster</h3>
      <p>Upload a simple product photo. Ask Happycapy to keep the object recognizable, then change the layout, lighting, props, and poster typography.</p>

      <div>
        <div>
          <img alt="Plain mug product photo before editing" />

          <div>Before</div>
        </div>

        <div>
          <img alt="Campaign poster after editing the mug product photo" />

          <div>After</div>
        </div>
      </div>

      <div>
        <p>Ask Happycapy</p>
        <code>Edit the uploaded product photo.Keep the mug recognizable, but turn it into a 4:5 product campaign poster. Add clear poster typography with the headline “MORNING RITUAL” and the small label “CERAMIC MUG”.Use warm studio lighting, a premium editorial layout, soft shadows, and simple lifestyle props. Save it as outputs/mug-campaign-poster.png.</code>
      </div>
    </div>
  </div>
</div>

### 3. Combine references

Upload several images and tell Happycapy what each one controls: brand mark, subject, material, mood, layout, or scene.

<div>
  <div>
    <div>
      <div>Multi-reference</div>
      <h3>Create a merch campaign from three references</h3>
      <p>Use one image for the brand mark, one for the merch objects, and one for the campaign mood. Tell Happycapy what each reference should control.</p>

      <div>
        <div>
          <div>
            <img alt="Happycapy logo reference" />

            <div>
              <div>Image 1</div>
              <p>Controls the Happycapy brand mark and line-icon identity.</p>
            </div>
          </div>

          <div>
            <img alt="Blank merchandise objects reference" />

            <div>
              <div>Image 2</div>
              <p>Controls the merch objects, materials, and product perspective.</p>
            </div>
          </div>

          <div>
            <img alt="Campaign layout and mood reference" />

            <div>
              <div>Image 3</div>
              <p>Controls the poster mood, color palette, light, and layout.</p>
            </div>
          </div>
        </div>

        <div>
          <img alt="Generated Happycapy merch campaign poster" />
        </div>
      </div>

      <div>
        <p>Ask Happycapy</p>
        <code>Use image 1 as the Happycapy brand mark reference. Keep the simple capybara line icon recognizable, but use it as a tasteful logo on the products.Use image 2 to control the merch objects, materials, and product perspective. Use image 3 to control the campaign mood, color palette, lighting, and layout.Create one 4:5 editorial campaign poster for a fictional Happycapy merch drop. Show the logo applied to the merch objects in a clean, premium way. Save it as outputs/happycapy-merch-campaign.png.</code>
      </div>
    </div>
  </div>
</div>

## Related resources

* [Files](/en/features/files): manage uploaded references, generated images, and `outputs/` files inside a Project.
* [AI Image Generator](https://happycapy.ai/tools/ai-image-generator): browse more image prompt ideas and try image generation directly.


# Models & Thinking Effort
Source: https://docs.happycapy.ai/en/features/models-thinking-effort

Choose a model, set effort when it is available, and control how deeply Happycapy works.

## What is Thinking Effort?

Happycapy gives you many models in the model selector. Some models also support **Effort**, which controls how much reasoning Happycapy should invest in the next task.

If a model supports effort control, Happycapy shows the **Effort** option automatically in the model menu. You do not need to memorize which models support it. Currently, supported models include Sonnet, Opus, GLM, and DeepSeek V4 Pro models, and this list may change as model availability changes.

Pick the model for the kind of work, then pick effort for how deeply Happycapy should work on it.

## Where to Find It

Open the model selector beside the message box. Choose a model. If the model supports effort control, Happycapy shows an **Effort** row under the model list.

<img alt="Happycapy model selector showing the Effort row and the Low, Medium, High, Extra, Max, and Adaptive thinking controls" />

The effort menu can include `Low`, `Medium`, `High`, `Extra`, `Max`, and **Adaptive thinking**.

Some models show a default label next to one effort level. Use that as the recommended starting point for that model.

You do not need to mention the effort level in your message. Set it in the model selector, then describe the task normally.

## Choose Before a Long Task

For quick questions, changing models or effort is usually fine. For a long or complex task, choose the model and effort before you start.

Try not to switch models or effort levels in the middle of the same complex task. Changing them can make Happycapy re-process more context, which may make the next step slower and use credits faster.

## How to Choose Effort

| Task                                                                               | Suggested effort | Why                                                   |
| ---------------------------------------------------------------------------------- | ---------------- | ----------------------------------------------------- |
| Rewrite one sentence, translate a paragraph, or answer a simple question           | `Low`            | Keeps the response fast and efficient.                |
| Summarize a short document or draft routine copy                                   | `Medium`         | Gives enough judgment without overthinking.           |
| Clean a CSV, analyze a file, or make a small code change                           | `High`           | Lets Happycapy inspect context and check the result.  |
| Debug code, compare options, or prepare a structured report                        | `Extra`          | Gives more room for planning, tool use, and review.   |
| Work across many files, produce a high-stakes deliverable, or solve a hard problem | `Max`            | Prioritizes quality over speed and credit efficiency. |

Start with the default effort when you are unsure. Move up when the task is complex, important, or expensive to redo. Move down when speed and credits matter more than depth.

## Use Adaptive Thinking

Adaptive thinking lets Happycapy spend less effort on simple requests and more effort on complex ones.

Use it when:

* the session mixes quick questions and deeper work
* you are not sure how complex the task will become
* you want Happycapy to decide when deeper reasoning is worth it

Adaptive thinking is still guided by the effort level you choose. A lower effort setting stays more conservative. A higher effort setting gives Happycapy more room to plan, inspect files, use tools, and check work.

## What to Know

Higher effort usually gives Happycapy more room to plan, inspect files, use tools, and review results. It also usually takes longer and can use credits faster.

Not every model supports effort control. If the model menu does not show **Effort**, use the model normally or choose another model that offers the setting.

## Related

* [Projects](/en/features/projects): keep related work and files together
* [Files](/en/features/files): give Happycapy source material to inspect
* [Dynamic Workflows](/en/features/dynamic-workflows): run larger tasks as staged work
* [Credits & Usage](/en/getting-started/credits): understand credit usage


# Projects
Source: https://docs.happycapy.ai/en/features/projects

Persistent workspaces for sessions, files, and ongoing work

## What is a Project?

A Project keeps related work in one workspace: sessions, files, and ongoing work. Use one Project for a codebase, campaign, client, research topic, content pipeline, or any task you expect to continue over time.

This makes Projects useful for work that spans more than one conversation. Start a task in one session, continue in another, and the files are already there.

## Start a Project

Click **New Project** in the sidebar, then describe what you want Happycapy to work on. Happycapy creates the Project and names it automatically based on your first prompt.

You can start with a prompt, upload files, or open an existing file in the Project.

<Tip>
  One Project per codebase or initiative is a clean setup. Use multiple sessions inside that Project when you want to run related tasks in parallel.
</Tip>

## Work across sessions

<div>
  <div>
    <div>
      <span>Sessions</span>
      <span>Parallel work, shared workspace.</span>
    </div>

    <img alt="Multiple sessions in a Project" />
  </div>

  <div>
    <p>Inside each Project, you can run multiple sessions at the same time. Each session is its own conversation thread, but they work in the same Project workspace.</p>
    <p>Practical uses:</p>

    <ul>
      <li>One session generates images while another writes the accompanying copy</li>
      <li>One session handles frontend code, another handles the backend</li>
      <li>One session researches a topic, another drafts the report</li>
    </ul>
  </div>
</div>

<p><strong>To open a new session:</strong> click the <strong>+</strong> button next to the session name at the top of the Project.</p>
<p><strong>To switch between sessions:</strong> click the session name dropdown at the top.</p>

## Keep files together

Sessions in the same Project can read and write the same files. Generated images, edited documents, code changes, reports, and other outputs stay with the Project instead of being scattered across separate conversations.

For more detail on browsing and managing project files, see [Files](/en/features/files).

## Folders and favorites

Folders group Projects in the sidebar. Click the **+** icon next to "Projects" (Create Folder) to make one.

Folders only change how Projects appear in the sidebar. They do not change the Project itself.

Click the **☆** icon in a session's header to pin it to the **Favorites** section at the top of the sidebar. Click again to remove it.


# Publish & Remix
Source: https://docs.happycapy.ai/en/features/publish-remix

Publish a remixable Project snapshot so others can copy it into a new Project, or so you can reuse it later.

## How Publish and Remix work

If you only want someone to view a conversation, file, or result, use **Share**.

Use **Publish** when you want someone to keep working from your Project. Publish saves a remixable Project snapshot. People with the link can view the snapshot, then **Remix** it into their own editable Project.

**Remix** creates a new Project from the snapshot. The copy is independent, so changes in the Remix do not affect the original Project.

You can also Remix your own published snapshot when you want to start from a stable version and explore another direction.

## When to use Publish and Remix

* You want someone to edit or adapt your Project
* You want a stable version that others can copy
* You want to explore another direction yourself
* You want a team or community to reuse your Project

<Note>
  Anyone with the published link can view the snapshot. Review files, chat history, tokens, secrets, and private information before you publish.
</Note>

## Publish a snapshot

Click **Publish** in the top-right corner of a Project.

<div>
  <div>
    <span>Publish</span>
    <span>Save a remixable Project snapshot.</span>
  </div>

  <img alt="Publish snapshot panel" />
</div>

Choose what to include before you copy the link:

* **Include chat history** — include the conversation in the snapshot
* **Files only** — turn off chat history if you only want to publish the Project files
* **Past snapshots** — find and reshare links you published earlier

After publishing, copy the link and send it to anyone who should view or Remix the snapshot.

<Tip>
  A snapshot captures the Project at the moment you publish it. If you make important changes later, publish again to share the newer version.
</Tip>

## Remix a snapshot

Open a published snapshot and click **Remix**.

<LivePreview label="happycapy.ai/snapshots/e4f0e3b2-a8c8-4c50-91df-0789f7017d37" />

Happycapy copies the snapshot into your workspace and opens it as a new Project. From there, you can edit files, continue the work, or ask Happycapy to take it in a new direction.

The copy is independent:

* changes in the Remix do not affect the original Project
* changes in the original Project do not update existing Remixes
* different people can Remix the same snapshot and work separately


# Skill Store
Source: https://docs.happycapy.ai/en/features/skill-store



## What is Skill Store?

Use the Skill Store to find and install ready-made Skills for design, coding, data analysis, file processing, media generation, and other work you repeat often.

Skills give Happycapy reusable instructions, context, or tools for specific kinds of work.

## How to use Skill Store

<div>
  <div>
    <span>Skill Store</span>
    <span>Browse, search, and install reusable Skills.</span>
  </div>

  <img alt="Skill Store" />
</div>

### Browse or search

Open the [Skill Store](https://happycapy.ai/skills) when you need a Skill for a task. Browse featured Skills, or search by task, tool, workflow, or domain.

### Install a Skill

Install a Skill once, then use it in your Happycapy Project. Installed Skills make repeated work easier because Happycapy can reuse the same workflow instead of starting from scratch each time.

### Use a Skill in a Project

Most of the time, you only need to describe the result you want. Happycapy will choose a relevant Skill if one is a good fit.

```text theme={null}
Create a product launch image for this landing page.
```

You can also choose a specific Skill from the picker or use a slash command when you want more control.

```text theme={null}
/frontend-design Design a landing page for a SaaS product.
```

## What to know

### Skills are reusable capabilities

A Skill is more than a prompt. It helps Happycapy follow a reusable workflow more consistently.

### Happycapy can choose Skills for you

You do not need to memorize Skill names. Describe the outcome you want, and Happycapy can select a Skill when one is a good fit.

### Add or create Skills for repeated work

Use Skills for work you repeat often. Start with the Skill Store to install a ready-made Skill. If the workflow is specific to your team or project, ask Happycapy to create a Skill for it.

## FAQ

### Do I need to choose a Skill manually?

No. In most cases, describe the result you want and Happycapy will choose a relevant Skill. Choose a Skill manually when you want a specific workflow or output style.

### Where can I find Skills?

Start with the [Skill Store](https://happycapy.ai/skills). You can also use compatible open-source Skills when you need a workflow that is not yet available in the store.


# Video Generation
Source: https://docs.happycapy.ai/en/features/video-generation

Create short videos from prompts, images, or existing clips inside a Project

## What is Video Generation?

Video Generation lets you create short videos from text prompts, images, or existing video clips. Use it for social ads, product demos, storyboards, motion concepts, and campaign assets you want to keep with the rest of your Project files.

Start from the material you already have. Click **Generate Video** below the input box to enter video mode, then describe the clip you want. You can also ask for a video directly in your message. If a model matters, choose it from the video model menu or mention the model in your prompt; otherwise, describe the result and let Happycapy choose a suitable path.

## Three ways to start

Choose the input you already have: a prompt, one or more images, or an existing video.

### 1. Start from text

Describe the video you want. Include the subject, scene, motion, style, duration, aspect ratio, and where you plan to use it.

<div>
  <div>
    <div>
      <div>Text to video</div>
      <h3>Turn a motion idea into a vertical clip</h3>
      <p>Start with the finished motion you need. Include the subject, camera angle, repeated action, sensory details, and target aspect ratio.</p>

      <div>
        <div>
          <p>Ask Happycapy</p>
          <code>Close-up shot of a hand pressing keys one by one on a keyboard made of assorted cookies and candies.Each time a fingertip lands, the corresponding cookie keycap slightly depresses, then quickly springs back up, making a crisp "click-clack, click-clack" sound that combines the crunchy texture of biscuits with the sharp tactile feedback of a mechanical keyboard.The fingers keep pressing different keycaps in a smooth, continuous motion with a steady rhythm. Vertical video, 9:16 aspect ratio.</code>
        </div>

        <div>
          <video>
            <source type="video/mp4" />

            Your browser does not support the video tag.
          </video>
        </div>
      </div>
    </div>
  </div>
</div>

### 2. Start from images

Upload images when you want more control over the first frame, last frame, subject, product, character, or visual style.

<div>
  <div>
    <div>
      <div>Image to video</div>
      <h3>Animate a product from reference images</h3>
      <p>Tell Happycapy what each image should control. A start frame sets the opening, an end frame sets the destination, and reference images guide identity, materials, or style.</p>

      <div>
        <div>
          <div>Start frame</div>
          <p>The first visible moment.</p>
        </div>

        <div>
          <div>End frame</div>
          <p>The final pose, layout, or reveal.</p>
        </div>

        <div>
          <div>Reference image</div>
          <p>The product, character, brand, or mood.</p>
        </div>
      </div>

      <div>
        <div>
          <img alt="Nine-panel lemon cider storyboard reference for an image-to-video ad" />

          <div>Reference storyboard</div>
        </div>

        <div>
          <video>
            <source type="video/mp4" />

            Your browser does not support the video tag.
          </video>

          <div>Generated video</div>
        </div>
      </div>

      <p>A storyboard can do two jobs at once: it anchors the product, character, and color palette, while also telling Happycapy the order of the shots.</p>

      <div>
        <p>Ask Happycapy</p>
        <code>Use the attached 9-panel storyboard image as the visual reference and shot sequence guide for a 15-second lemon sparkling water advertisement.Keep the character, product label, color palette, bright summer lighting, and camera compositions close to the reference. Follow the panels from left to right, top to bottom. Add natural motion: bottle condensation, rising bubbles, hair moving in the breeze, slow sipping, sunlight flares, and a final close-up freeze frame.Use an airy Japanese summer commercial style with vivid sky blue, warm golden sunlight, shallow depth of field, and smooth handheld camera motion. Save the result as outputs/lemon-cider-ad.mp4.</code>
      </div>
    </div>
  </div>
</div>

<Accordion title="Full storyboard prompt">
  <div>
    <code>Use the attached 9-panel storyboard image ([https://storage.res-capy.com/img-prompt/lemon-cider-ad.png](https://storage.res-capy.com/img-prompt/lemon-cider-ad.png)) as the visual reference and shot sequence guide for this advertisement video.The storyboard shows a cheerful Japanese young woman in a white outfit holding a transparent lemon sparkling water bottle ("レモンサイダー LEMON CIDER 瀬戸内レモン") against a bright blue summer sky. Reproduce the character's appearance, product design, color palette, and each shot composition faithfully from the reference.Follow the 9 panels in order as the scene sequence:Panel 1 (0:00-0:02): Animate the bottle being thrust toward camera, condensation droplets catching the light, bubbles rising inside the bottle.Panel 2 (0:02-0:04): The girl turns to face camera with a bright smile, her hair moves gently in the breeze, sunlight glints behind her.Panel 3 (0:04-0:06): She slowly tilts her head back to take a sip, eyes closing in satisfaction, a warm sunlight halo blooms around her silhouette.Panel 4 (0:06-0:08): She walks along the sun-drenched path, glances back over her shoulder with a playful smile, leaves shimmer in the background.Panel 5 (0:08-0:10): The bottle floats centered in frame against the deep blue sky, sunlight refracts through it creating a slow rainbow arc sweep.Panel 6 (0:10-0:11): She extends the bottle toward the viewer with both hands, leaning slightly forward with a warm inviting smile.Panel 7 (0:11-0:12): Extreme close-up: the bottle cap twists open in slow motion, fizz and bubbles burst upward backlit by golden sunlight.Panel 8 (0:12-0:13): She raises the bottle high above her head laughing freely, camera drifts upward to reveal the vast open blue sky.Panel 9 (0:13-0:15): She holds the bottle close to her face, looks softly into the camera with a genuine smile. Slow push in. Freeze on final frame.STYLE: Match the color grading in the reference exactly: overexposed highlights, airy whites, vivid sky blue, warm golden sunlight. Cinematic shallow depth of field. Smooth handheld camera motion with subtle drift. Slow motion on sip, fizz, and final freeze moments.</code>
  </div>
</Accordion>

### 3. Edit a video

Upload a clip when you already have footage and want Happycapy to create a new version, change the subject, adjust the action, or keep refining the same scene.

<div>
  <div>
    <div>
      <div>Video editing</div>
      <h3>Edit what happens in an existing clip</h3>
      <p>Give Happycapy the source video and describe what should change. You can keep the original camera movement, room, lighting, and timing while changing the subject or action.</p>

      <div>
        <div>
          <video>
            <source type="video/mp4" />

            Your browser does not support the video tag.
          </video>

          <div>Before</div>
        </div>

        <div>
          <video>
            <source type="video/mp4" />

            Your browser does not support the video tag.
          </video>

          <div>After</div>
        </div>
      </div>

      <div>
        <p>Ask Happycapy</p>
        <code>Use the uploaded video as the source.Keep the original room, floor, desk legs, camera movement, lighting, and the cat's position. Add a capybara into the scene so it comes out from the laptop screen, lands naturally on the floor, and has a playful scuffle with the cat.Make the capybara match the perspective, scale, shadows, and handheld phone-video style. Keep the interaction energetic but not violent. Save the edited video as outputs/capybara-cat-edit.mp4.</code>
      </div>
    </div>
  </div>
</div>

## Keep iterating

After a video is generated, keep working in the same conversation. Ask Happycapy to make another version, adjust the motion, create a cover image, write a caption, or save the final files with clear names.

## Related resources

* [Image Creation & Editing](/en/features/image-generation): create still images, references, and cover assets.
* [Files](/en/features/files): find generated MP4s, cover images, and supporting files.
* [Credits & usage](/en/getting-started/credits): understand how media generation uses credits.
* [Generate a short video ad](/en/use-cases/short-video-ad): see a complete video ad workflow.


# Credits & usage
Source: https://docs.happycapy.ai/en/getting-started/credits

Understand credits, usage charts, model choice, and plan limits in Happycapy

Credits are the shared usage unit for work that runs in Happycapy. Model calls, media generation, long-running tasks, and Automations can all use credits.

Use this page to understand where your credits go and how to make better choices when a task can be done in more than one way.

<Note>
  Plan details can change. For current prices, plan limits, and billing terms, use the official [Happycapy pricing page](https://happycapy.ai/pricing).
</Note>

## Check usage

Open the profile menu in the lower-left corner, choose **Settings**, then open **Billing**. In the **Usage** section, you will see:

* **Credits consumed** — how many credits were used in the selected time range.
* **Requests made** — how many model or tool requests were made in the selected time range.

Use **7 Days**, **30 Days**, or **90 Days** to switch the reporting window.

## Read the charts

<CardGroup>
  <Card title="Credits by Model" icon="coins">
    Shows daily credit usage for the models used in the selected time range. Use it to compare which models used more credits on each day.
  </Card>

  <Card title="Requests by Model" icon="chart-column">
    Shows request volume instead of cost. A lightweight model may appear often here while using fewer credits overall.
  </Card>
</CardGroup>

Look at both charts together. A model can have many requests but low credit usage, or fewer requests but higher credit usage.

## What uses more credits

Credit usage depends on what Happycapy is doing, not just how many messages you send.

<CardGroup>
  <Card title="Model choice" icon="sliders">
    Higher-capability reasoning models usually use more credits than faster, lighter models.
  </Card>

  <Card title="Task length" icon="route">
    Multi-step research, coding, file processing, and revision loops can use more credits as Happycapy works through the task.
  </Card>

  <Card title="Media generation" icon="palette">
    Skills can trigger specialized image or video models. Those credits are tracked separately from base-model chat work.
  </Card>

  <Card title="Automations" icon="clock">
    Scheduled tasks can keep running even when you are away, so review usage regularly if you automate recurring work.
  </Card>
</CardGroup>

## Choose models

The model selector is for the base model that drives your Happycapy session.

<CardGroup>
  <Card title="Claude" icon="https://mintcdn.com/happycapy/9SccTkPwiYpehlgU/images/models/logos/claude.svg?fit=max&auto=format&n=9SccTkPwiYpehlgU&q=85&s=d497305833cfb31f61feb6b91f4291fb">
    Opus for complex reasoning and long-horizon coding, Sonnet for balanced everyday work, Haiku for fast lighter tasks.
  </Card>

  <Card title="MiniMax" icon="https://mintcdn.com/happycapy/9SccTkPwiYpehlgU/images/models/logos/minimax.svg?fit=max&auto=format&n=9SccTkPwiYpehlgU&q=85&s=0b66f0676774eab05151cd73f7df5346">
    M3 for coding, agentic workflows, long-context tasks, and multimodal inputs.
  </Card>

  <Card title="DeepSeek" icon="https://mintcdn.com/happycapy/9SccTkPwiYpehlgU/images/models/logos/deepseek.svg?fit=max&auto=format&n=9SccTkPwiYpehlgU&q=85&s=6b6e6e0cebdcf256dd05465615388554">
    V4 Pro for advanced reasoning and agentic coding, V4 Flash for fast and efficient work.
  </Card>

  <Card title="GLM" icon="https://mintcdn.com/happycapy/qu9DbeP3oO1Bbe3K/images/models/logos/glm.svg?fit=max&auto=format&n=qu9DbeP3oO1Bbe3K&q=85&s=74e15d4d2d55fe9744cd6521024f1d60">
    5.2 for coding and long tasks when you want capable agent work with fewer credits.
  </Card>

  <Card title="Kimi" icon="https://mintcdn.com/happycapy/qu9DbeP3oO1Bbe3K/images/models/logos/kimi.svg?fit=max&auto=format&n=qu9DbeP3oO1Bbe3K&q=85&s=3117f3f523d91c8afee40ec986be399a">
    K2.7 Code for coding-focused tasks; K2.6 for multimodal inputs and longer execution.
  </Card>
</CardGroup>

Specialized media models such as GPT Image 2, Gemini, and Seedream are task-level models, not base models in the regular selector. You can ask for them in the conversation or use a Skill that calls them; Happycapy will route the image or video task to the right model.

If you are unsure, start with the default base model. Switch to a stronger model when the task needs deeper reasoning, or switch to a lighter model for quick edits and repeated small steps.

## Plans

Happycapy has Free, Pro, Plus, Max, and Team plans. Annual billing on the pricing page includes two months free.

<div>
  <table>
    <colgroup>
      <col />

      <col />

      <col />

      <col />
    </colgroup>

    <thead>
      <tr>
        <th>Plan</th>
        <th>Credits / month</th>
        <th>Sandbox</th>
        <th>Good for</th>
      </tr>
    </thead>

    <tbody>
      <tr>
        <td>Free</td>
        <td>250</td>
        <td>Basic sandbox environment</td>
        <td>Trying Happycapy and small tasks</td>
      </tr>

      <tr>
        <td>Pro</td>
        <td>2,000</td>
        <td>2 cores, 4GB RAM, 50GB storage</td>
        <td>Everyday productivity and recurring tasks</td>
      </tr>

      <tr>
        <td>Plus</td>
        <td>5,000</td>
        <td>2 cores, 4GB RAM, 50GB storage</td>
        <td>Heavier personal use</td>
      </tr>

      <tr>
        <td>Max</td>
        <td>22,000</td>
        <td>4 cores, 8GB RAM, 200GB storage</td>
        <td>Ambitious projects, more Automations, and higher usage</td>
      </tr>
    </tbody>
  </table>
</div>

<Card title="View current pricing" icon="arrow-up-right" href="https://happycapy.ai/pricing">
  See current prices, annual billing, included Automations, Capy Mail quota, and Team plan details.
</Card>


# Frequently asked questions
Source: https://docs.happycapy.ai/en/getting-started/faq

Answers to common questions about Happycapy

## Common questions

<AccordionGroup>
  <Accordion icon="https://mintcdn.com/happycapy/9SccTkPwiYpehlgU/images/icons/lucide-shield-check.svg?fit=max&auto=format&n=9SccTkPwiYpehlgU&q=85&s=9248d70170b8d24e1edeab5295f2cee3" title="Is my data safe?">
    Happycapy runs tasks in a cloud sandbox that is separate from your local computer. Happycapy can only use files you upload or add to a Project.

    You authorize every connection to third-party tools, and you can revoke access from the connected service at any time.

    See the [Privacy Policy](https://happycapy.ai/privacy) for the latest data and privacy terms.
  </Accordion>

  <Accordion icon="https://mintcdn.com/happycapy/9SccTkPwiYpehlgU/images/icons/lucide-terminal.svg?fit=max&auto=format&n=9SccTkPwiYpehlgU&q=85&s=c89791fab377e5bc8a40d27a087689ee" title="Do I need to know how to code?">
    No. Describe what you want in natural language, whether the task is writing, research, processing files, generating images, or building an app.

    If you do know how to code, you can give technical instructions, request specific frameworks, and review generated files directly.
  </Accordion>

  <Accordion icon="https://mintcdn.com/happycapy/9SccTkPwiYpehlgU/images/icons/lucide-files.svg?fit=max&auto=format&n=9SccTkPwiYpehlgU&q=85&s=f6c948280d9796322d01603d044e4c26" title="What kinds of files can Happycapy work with?">
    Happycapy can work with common formats, including documents, spreadsheets, PDFs, images, code, CSV, JSON, audio, and video.

    Use [Files](/en/features/files) to manage uploads and generated outputs.
  </Accordion>

  <Accordion icon="https://mintcdn.com/happycapy/9SccTkPwiYpehlgU/images/icons/lucide-brain-circuit.svg?fit=max&auto=format&n=9SccTkPwiYpehlgU&q=85&s=dd9c5b2a9c9e8d18c0d2a491e8ac7a0f" title="Which models does Happycapy support?">
    Happycapy supports base model families such as Claude, MiniMax, DeepSeek, and Kimi. For media tasks, Happycapy can also route to task-level models such as GPT Image 2, Gemini, or Seedream.

    See [Credits & Usage](/en/getting-started/credits) for the latest model overview.
  </Accordion>

  <Accordion icon="https://mintcdn.com/happycapy/9SccTkPwiYpehlgU/images/icons/lucide-coins.svg?fit=max&auto=format&n=9SccTkPwiYpehlgU&q=85&s=64b1365b14e8bbe1cb3422a53de01ca6" title="How do credits and plans work?">
    Credits track usage across model calls, media generation, long-running tasks, and Automations. Plan details can change, so the latest pricing lives on the official website.

    Read [Credits & Usage](/en/getting-started/credits), or view current plans on [happycapy.ai/pricing](https://happycapy.ai/pricing).
  </Accordion>

  <Accordion icon="https://mintcdn.com/happycapy/9SccTkPwiYpehlgU/images/icons/lucide-smartphone.svg?fit=max&auto=format&n=9SccTkPwiYpehlgU&q=85&s=cca977186c97d529bd780be4e9950d71" title="Can I use Happycapy on mobile?">
    Yes. Happycapy works in the browser and as an iOS or Android app. Search for "Happycapy" in the App Store or Google Play.

    Larger screens are usually better for complex file work, code review, and multi-step Projects.
  </Accordion>

  <Accordion icon="https://mintcdn.com/happycapy/9SccTkPwiYpehlgU/images/icons/lucide-plug.svg?fit=max&auto=format&n=9SccTkPwiYpehlgU&q=85&s=47501e8e870baabd9ae9479a383367fc" title="Can Happycapy connect to other tools?">
    Yes. Happycapy can work with other tools and systems through CLI tools, MCP servers, APIs, and user-authorized connections.

    Start with [Integrations](/en/integrations/cli) to learn how CLI tools and MCP servers can help Happycapy work with outside and internal systems.
  </Accordion>

  <Accordion icon="https://mintcdn.com/happycapy/9SccTkPwiYpehlgU/images/icons/lucide-life-buoy.svg?fit=max&auto=format&n=9SccTkPwiYpehlgU&q=85&s=8ac474f56d85b0b7816c1d6b7eb8c83b" title="Where can I get help?">
    You can browse these docs, join [Discord](https://discord.gg/kSwMyUCdwH), send [product feedback](https://feedback.happycapy.ai), or email [support@happycapy.ai](mailto:support@happycapy.ai).
  </Accordion>
</AccordionGroup>


# Quick Start
Source: https://docs.happycapy.ai/en/getting-started/quickstart

Finish your first task in Happycapy in a few minutes

## 1. Open Happycapy

Visit [happycapy.ai](https://happycapy.ai), sign in, and open your first Project.

<div>
  <div>
    <span />

    <span />

    <span />

    <span>happycapy.ai</span>
  </div>

  <img alt="Happycapy main interface" />
</div>

* **Sidebar** — Create Projects, browse the Skill Store, and set up Automations.
* **Projects** — Work with Happycapy in a persistent workspace and follow progress as it runs.
* **Input box** — Type a request, attach files with **+**, choose a Skill, or pick a model when needed.

## 2. Ask for a result

Tell Happycapy what you want done. A good request usually includes the goal, context, and desired output.

```text theme={null}
Build a personal portfolio website with a dark theme and a contact form.
```

```text theme={null}
Analyze this CSV and create a chart showing monthly revenue trends.
```

```text theme={null}
Create a product launch image for this landing page.
```

Happycapy works in a cloud sandbox. Each Project has its own workspace inside that sandbox, where generated files and outputs stay. Open **Files** when you want to review, download, or reuse them.

## 3. Add files when the task needs context

Drag files into the conversation, or click **+** in the input box.

Happycapy can work with common files such as spreadsheets, PDFs, images, code, videos, CSV, and JSON.

Use files when you want Happycapy to analyze source material, transform an existing asset, or continue work from something you already have.

If a file or folder is already in the Project, type `@` in the input box and choose it from **Session Files**. This works for uploaded files, folders, and files Happycapy generated under `outputs/`, so you can ask Happycapy to inspect, edit, summarize, convert, or continue from a specific artifact without uploading it again.

<img alt="Mention Project files and folders from Session Files with @" />

## 4. Use Skills for repeated work

You do not need to install a Skill for every task. Describe what you want first, and Happycapy can choose relevant Skills when they fit the work.

Use the [Skill Store](/en/features/skill-store) when you want a repeatable workflow or a specific capability. You can also browse public [Skills](https://happycapy.ai/skills) on the website.

You can also type `/` in the input box to use slash commands. Slash commands can launch Skills directly, which is faster when you already know the workflow you want.

```text theme={null}
/frontend-design Create a landing page for a SaaS product.
```

## 5. Keep going

<CardGroup>
  <Card title="Projects" icon="desktop" href="/en/features/projects">
    Use Projects as persistent workspaces for sessions, files, and ongoing work.
  </Card>

  <Card title="Files" icon="file" href="/en/features/files">
    Find generated outputs, uploaded files, and materials you want to reuse.
  </Card>

  <Card title="Skill Store" icon="wand-magic-sparkles" href="/en/features/skill-store">
    Add reusable Skills for repeated workflows and specific capabilities.
  </Card>

  <Card title="Automations" icon="clock" href="/en/features/automations">
    Schedule recurring tasks and long-running work from a Project.
  </Card>
</CardGroup>


# Welcome to Happycapy
Source: https://docs.happycapy.ai/en/index



## What is Happycapy?

Happycapy is an agent-native computer that runs in your browser.

No installs. No terminal setup. No API keys. Open a tab and use Happycapy for getting work done.

Happycapy can write code, process documents, generate images, analyze data, and run tasks on your behalf.

Happycapy is powered by Claude Code and routes each task to suitable built-in models for the scenario, so you can focus on the outcome instead of managing model choices.

## What makes Happycapy different?

<CardGroup>
  <Card title="Agent-native work" icon="sparkles">
    Break down tasks, use tools, execute steps, and report back with results.
  </Card>

  <Card title="Cloud sandbox" icon="cloud" href="/en/features/projects">
    Run work in the cloud from a browser while your local files stay separate.
  </Card>

  <Card title="Skill Store" icon="wand-magic-sparkles" href="/en/features/skill-store">
    Use purpose-built Skills for generation, coding, analysis, translation, and more.
  </Card>

  <Card title="Automations" icon="clock" href="/en/features/automations">
    Schedule recurring and long-running work that continues in the cloud.
  </Card>
</CardGroup>

## Watch the intro

<div>
  <iframe title="Introducing Happycapy" />
</div>

## Start here

<CardGroup>
  <Card title="Quick Start" icon="rocket" href="/en/getting-started/quickstart">
    Learn the basic Happycapy workflow in a few minutes.
  </Card>

  <Card title="Credits & Usage" icon="coins" href="/en/getting-started/credits">
    Understand credits, model usage, and how to review your usage.
  </Card>

  <Card title="Skill Store" icon="wand-magic-sparkles" href="/en/features/skill-store">
    Browse ready-made Skills and learn how to use them in Projects.
  </Card>

  <Card title="Automations" icon="clock" href="/en/features/automations">
    Schedule work that keeps running even when you are away.
  </Card>
</CardGroup>


# CLI tools
Source: https://docs.happycapy.ai/en/integrations/cli

Run command-line tools in the Happycapy sandbox.

## Overview

A command-line interface, or CLI, is a tool you run with commands instead of buttons. Many services provide CLIs so users can manage repositories, documents, deployments, logs, and workspace data programmatically.

In Happycapy, you do not need to memorize those commands. Describe what you want to do in natural language, and Happycapy can install the CLI, choose the right commands, run them in the sandbox, read the output, and continue the work inside your Project.

A CLI may connect to an external service such as GitHub, Notion, Lark, or Fly.io. It may also be a local utility for working with files, code, media, or data inside the sandbox.

## When to use CLI tools

Use CLI tools when a service exposes useful actions through commands:

* Work with repositories, pull requests, issues, logs, and deployments
* Read or update workspace systems such as Notion and Lark
* Run cloud, database, or infrastructure commands
* Turn repeatable command-line work into a Project workflow

## CLI or MCP?

CLI and MCP can feel similar in Happycapy: you describe the goal in natural language, and the agent decides which tool calls or commands are needed.

The difference is the connection layer. With CLI tools, Happycapy writes and runs command-line programs in the sandbox. With [MCP](/en/integrations/mcp), Happycapy calls tools exposed by an MCP server.

If a service has a well-maintained official CLI, or the task involves deployments, logs, scripts, files, or one-off setup, CLI is often the direct path. If a service has a maintained MCP server, or you want tool actions exposed through a structured interface, MCP may fit better. If both exist, choose the option that is officially maintained, clear about permissions, and easiest to configure.

## Before you start

Helpful context to provide:

* The official CLI name or install link
* The account, workspace, repository, or app to connect
* The required auth method, such as login flow, token, API key, or workspace permission
* The outcome you want Happycapy to complete after setup

## First use

Give Happycapy the tool and the goal:

```text theme={null}
Install [CLI name] in the Happycapy sandbox.
Tell me what authentication is required before you continue.
After it is ready, use it to [goal].

CLI link: [official link]
```

## Recommended CLIs

These CLIs work well in a Happycapy sandbox because they expose real systems through commands that an agent can install, configure, and run.

<CardGroup>
  <Card title="GitHub CLI" icon="https://mintcdn.com/happycapy/9SccTkPwiYpehlgU/images/integrations/logos/github.png?fit=max&auto=format&n=9SccTkPwiYpehlgU&q=85&s=679bb093797d36e654af372f853a460a" href="https://cli.github.com">
    Work with repositories, pull requests, issues, and GitHub Actions.
  </Card>

  <Card title="ntn" icon="https://mintcdn.com/happycapy/9SccTkPwiYpehlgU/images/integrations/logos/notion.png?fit=max&auto=format&n=9SccTkPwiYpehlgU&q=85&s=534c589e6711180fe51fa3b632720e51" href="https://developers.notion.com/cli">
    Search Notion pages, create content, and query databases.
  </Card>

  <Card title="lark-cli" icon="https://mintcdn.com/happycapy/9SccTkPwiYpehlgU/images/integrations/logos/lark.png?fit=max&auto=format&n=9SccTkPwiYpehlgU&q=85&s=adc21fbe2f882fafdd5006135fb05128" href="https://github.com/larksuite/cli">
    Work with Lark/Feishu messages, docs, sheets, calendar, tasks, mail, and meetings.
  </Card>

  <Card title="flyctl" icon="https://mintcdn.com/happycapy/9SccTkPwiYpehlgU/images/integrations/logos/flyio.svg?fit=max&auto=format&n=9SccTkPwiYpehlgU&q=85&s=d9342a993c06f2170919e5f5a526a3d3" href="https://fly.io/docs/flyctl/">
    Deploy and operate Fly.io apps, Machines, logs, secrets, and releases.
  </Card>
</CardGroup>

## Example workflows

```text theme={null}
Use GitHub CLI.
Review the latest failed GitHub Actions run for this repository.
Summarize the failure and suggest the smallest fix.
```

```text theme={null}
Use ntn.
Search our Notion workspace for onboarding docs.
Summarize the current process and list gaps.
```

```text theme={null}
Use lark-cli.
Collect this week's project updates from Lark docs.
Turn them into a short team report.
```

```text theme={null}
Use flyctl.
Check the status and recent logs for this Fly.io app.
Tell me whether the latest deploy is healthy.
```

## Authentication and permissions

* Start with the account, workspace, repository, or project the CLI should connect to.
* Keep secrets out of prompts when possible. Prefer the CLI's normal authentication flow.
* Ask Happycapy to explain required permissions before you approve authentication.
* For tokens, prefer scoped or short-lived tokens when the service supports them.
* For destructive operations, ask Happycapy to show the command and expected effect before running it.

## Troubleshooting

### The CLI cannot install

Share the official install link and ask Happycapy to check the sandbox environment, package manager, and platform requirements.

### Login failed

Confirm whether the CLI expects browser login, token login, workspace approval, or environment variables.

### Permission denied

Ask Happycapy to show which permission is missing and whether the action needs a different token, role, workspace, repository, or organization.

### The output is too broad

Narrow the request with a repository, project, date range, file path, app name, or command target.

## Next steps

* Read [MCP](/en/integrations/mcp) to see another way to connect tools. MCP feels similar to CLI in use, but calls tools from an MCP server instead of running command-line programs in the sandbox.
* Use [Automations](/en/features/automations) when the same CLI workflow should run on a schedule.


# Custom Secrets
Source: https://docs.happycapy.ai/en/integrations/custom-secrets

Save private access keys so Happycapy can use external tools without exposing them in chat.

Custom secrets are private access keys for services outside Happycapy, such as GitHub, Notion, Fly.io, Stripe, or your own internal tools.

When a tool needs a token or API key, add it in **Settings > Sandbox > Custom secrets** instead of pasting it into chat. Happycapy injects the secret into the sandbox as an environment variable on each message.

## When to Use Custom Secrets

Use custom secrets when Happycapy needs to access a service or tool on your behalf.

Common examples include:

* checking GitHub repositories, pull requests, or workflow runs
* creating or updating Notion pages
* deploying or inspecting apps on platforms like Fly.io
* running scripts that call your own Stripe, analytics, database, or internal APIs
* using command-line tools that expect credentials from environment variables

## Add a Secret

1. Open **Settings**.
2. Go to **Sandbox**.
3. Find **Custom secrets**.
4. Click **Add**.
5. Enter a **Key**, **Value**, and **Description**.
6. Click **Create**.

<Frame>
  <img alt="Custom secrets settings in the Happycapy sandbox settings panel" />
</Frame>

## What Each Field Means

| Field           | Plain meaning                                             | Example                                                  |
| --------------- | --------------------------------------------------------- | -------------------------------------------------------- |
| **Key**         | The name Happycapy and tools use to find this secret.     | `GITHUB_TOKEN`                                           |
| **Value**       | The actual private token, API key, or secret value.       | Paste the value from GitHub, Notion, or another service. |
| **Description** | A short note that tells Happycapy what the secret is for. | `GitHub access for checking PRs and workflow runs.`      |

Write descriptions as short instructions, not just labels.

Good:

```text theme={null}
GitHub token for using the gh CLI and GitHub API in my repositories.
```

Too vague:

```text theme={null}
github token
```

## Common Examples

| Key                  | Use it for                                                                           |
| -------------------- | ------------------------------------------------------------------------------------ |
| `GITHUB_TOKEN`       | Reading repositories, checking pull requests, using the GitHub API, or running `gh`. |
| `NOTION_TOKEN`       | Working with Notion pages or databases through the Notion API.                       |
| `FLY_API_TOKEN`      | Deploying or checking Fly.io apps with `flyctl`.                                     |
| `STRIPE_API_KEY`     | Testing scripts that call your own Stripe account.                                   |
| `INTERNAL_API_TOKEN` | Calling your company or project API from a script.                                   |

These names are examples. Always use the key name expected by the tool, SDK, or CLI you plan to use.

## How Happycapy Uses Secrets

After you add a custom secret, you do not need to paste the value into chat. Ask for the task normally.

```text theme={null}
Use GitHub to check the latest failed workflow run and summarize what broke.
```

```text theme={null}
Use the Notion API to create a task list from this meeting summary.
```

Happycapy can see that the environment variable exists and can use your description to understand what it is for. Secret values stay hidden in the interface.

## Tips

Do not paste private tokens directly into chat.

Use clear, standard key names when possible. Tools often look for exact environment variable names.

Keep descriptions specific. Mention the service, tool, and intended use.

Delete secrets you no longer need.

If a tool cannot find a secret, check that the key name matches what the tool expects.

## Related resources

* [CLI tools](/en/integrations/cli): use command-line tools that read credentials from environment variables.
* [MCP](/en/integrations/mcp): connect Happycapy to external tools and systems through MCP servers.
* [Files](/en/features/files): manage uploaded files and generated outputs in the sandbox.


# MCP
Source: https://docs.happycapy.ai/en/integrations/mcp

Use MCP servers as integrations in Happycapy.

## Overview

MCP lets Happycapy use another set of tools and data inside a Project. You can ask it to look up context, read records, update systems, query databases, or run team-specific actions.

You do not need to understand the protocol. Give Happycapy the server setup guide, access details, and goal. Happycapy can inspect what the server provides, explain what access is needed, and use those tools while it works.

You can use MCP with public apps like Gmail and Notion, or with internal knowledge bases, databases, CRMs, approval flows, dashboards, operations tools, and legacy APIs.

## When to use MCP

Use MCP when you want Happycapy to work with a system instead of only working with the files and context already in the session:

* Search docs, pages, messages, or team knowledge across tools like Notion, Google Drive, Slack, or an internal wiki
* Read or update email, calendar, CRM, support, ticketing, or project systems
* Query databases and internal data systems such as Postgres, Supabase, Neon, dashboards, or reporting tools
* Work with developer and product systems such as GitHub, Linear, logs, issues, or release data
* Use team-built actions such as approvals, inventory lookup, customer record updates, or report generation

Whether a tool can be used depends on whether it has an MCP server or setup guide you can provide to Happycapy.

## What MCP can help with

<CardGroup>
  <Card title="Bring in context" icon="search">
    Let Happycapy find relevant docs, tickets, messages, records, or customer notes before it starts the work.
  </Card>

  <Card title="Use approved actions" icon="check-circle">
    Ask Happycapy to draft, update, create, assign, comment, sync, or summarize through tools exposed by the server.
  </Card>

  <Card title="Work with internal systems" icon="building">
    Connect team-built servers for internal databases, dashboards, approvals, operations, or legacy systems.
  </Card>

  <Card title="Reuse a setup" icon="repeat">
    Keep the server details in a Project so similar workflows can use the same connection later.
  </Card>
</CardGroup>

## Before you start

Helpful details:

* The MCP server link, repository, or setup guide
* The account, workspace, project, or data source it should use
* Any login, token, permission, or workspace approval instructions
* The task you want Happycapy to complete after the server is ready

## First use

Give Happycapy the server and the goal:

```text theme={null}
Set up this MCP server for this Happycapy Project.
Before making changes, tell me what access it needs.
After it is ready, show me the tools it exposes.

MCP server: [link or setup guide]
Goal: [what you want Happycapy to do]
```

## Example workflows

```text theme={null}
Use our Notion MCP server.
Find the latest product launch notes and summarize the open decisions.
```

```text theme={null}
Use the Gmail and Calendar MCP server.
Find customer meeting emails from this week and draft a follow-up plan.
```

```text theme={null}
Use the internal CRM MCP server.
Look up this customer, summarize recent activity, and draft an account update.
```

```text theme={null}
Use the Postgres MCP server.
Query last week's activation data and explain the trend in plain English.
```

## Access and permissions

* Prefer the server's official login or authorization flow.
* Use the smallest permission set that can complete the task.
* Ask Happycapy to list available tools before it starts making changes.
* For sensitive actions, ask Happycapy to explain what it will read or change before it runs the tool.
* Confirm the right account and workspace if the server can access company or customer data.

## Troubleshooting

### Setup does not work

Share the official setup guide and ask Happycapy to check what step failed.

### Authentication fails

Check whether the server expects a browser login, token, API key, environment variable, or workspace approval.

### The expected tool is missing

Ask Happycapy to list the tools the server exposes, then compare that list with the server docs.

### The result is too broad

Add the workspace, project, data source, record, date range, or output format you expected.

## For developers

MCP stands for Model Context Protocol. If you are building or choosing a server, use the official MCP documentation as the source of truth.

<Card title="Model Context Protocol" icon="globe" href="https://modelcontextprotocol.io">
  Read the official MCP documentation and server resources.
</Card>

## Next steps

* Keep the server link, access notes, and workspace details in your [Project](/en/features/projects) if you plan to reuse the setup.
* Use [Automations](/en/features/automations) when the same MCP workflow should run on a schedule.