[Integrations](https://spiral.ai/integrations) › [OpenClaw](https://spiral.ai/integrations/openclaw) › [SAM 3 Image](https://spiral.ai/integrations/openclaw/sam-3-image)

# How to Use SAM 3 Image in OpenClaw

SAM 3 Image finds and masks objects in existing photos using short descriptions or visual guidance. Use it to prepare product cutouts, editing selections, and annotation drafts.

|  |  |
| --- | --- |
| Model | SAM 3 Image |
| Available through | fal |
| Category | Image |
| Price | $0.0065 per run at the default inputs. |
| Returns | image |
| Runs | In the background |
| Setup | About 2 minutes, no API key |

## Prompts to try with SAM 3 Image in OpenClaw

- Each morning, use SAM 3 Image to select 'table lamp' in newly submitted product photos. Prepare a review gallery of the selections before cutout editing.
- Every afternoon, use SAM 3 Image to find 'chair' in new room survey photos. Report proposed counts with marked previews for review.
- Each day, use SAM 3 Image to draft 'dog' masks in newly submitted training images. Send the selections to the annotation review queue.
- Each evening, use SAM 3 Image to locate 'license plate' in new vehicle photos. Queue marked previews for review before the redaction step.
- Each morning, use SAM 3 Image to select 'bird' in new camera-trap stills. Prepare the masks for review and leave species identification to the reviewer.

## How to set up SAM 3 Image in OpenClaw

Pick the way that suits you. Each takes about two minutes.

### Use the Spiral skill

1. **Send OpenClaw this message**: Your agent reads Spiral's skill and sets itself up.

```
set up https://spiral.ai/SKILL.md
```

2. **Sign in or paste a key**: Approve Spiral in the browser, or give the agent a key from the console.

### Use Spiral's MCP server

1. **Add the server**: On the machine that runs OpenClaw's Gateway:

```bash
openclaw mcp add spiral \
  --url https://api.spiral.ai/mcp \
  --transport streamable-http \
  --auth oauth
```

2. **Sign in**: In Settings › MCP, select Sign in on the Spiral server, sign in to Spiral and click Authorize. No API key.

```bash
openclaw mcp login spiral
```

3. **Reload and check**: A running Gateway may keep its old tool list: restart it, or Save & Publish in the Control UI.

```bash
openclaw mcp reload
openclaw mcp doctor spiral --probe
```

### Use the Spiral CLI

1. **Install and sign in**: Installs the Spiral CLI, signs you in and adds its skill to your coding agents.

```bash
npx -y spiral-cli@latest init
```

### Call the Spiral API

1. **Create an API key**: In the console, under API keys.

2. **Run SAM 3 Image**: One request; the result comes back by polling or a webhook.

```bash
curl https://api.spiral.ai/v1/runs \
  -H "Authorization: Bearer $SPIRAL_API_KEY" \
  -H "Content-Type: application/json" \
  -H "Idempotency-Key: $(uuidgen)" \
  -d '{
    "provider": "fal",
    "model": "fal-ai/sam-3/image",
    "input": {
      "image_url": "https://raw.githubusercontent.com/facebookresearch/segment-anything-2/main/notebooks/images/truck.jpg"
    }
  }'
```

Then paste one of the prompts above: OpenClaw finds SAM 3 Image on Spiral, shows the price and runs it.

Setup guide in the docs: https://spiral.ai/docs/quickstarts/openclaw

## SAM 3 Image tools and prices

- [Segment Anything Model 3 · Image to Image](https://spiral.ai/tools/fal/fal-ai/sam-3/image) (fal): $0.0065 per run at the default inputs.

## Settings you can ask for

| Setting | Allowed values | Default | Description |
| --- | --- | --- | --- |
| prompt |  | wheel | Text prompt for segmentation |
| text_prompt |  | — | [DEPRECATED] Use 'prompt' instead. Kept for backward compatibility. |
| point_prompts |  | [] | List of point prompts |
| box_prompts |  | [] | Box prompt coordinates (x_min, y_min, x_max, y_max). Multiple boxes supported - use object_id to group boxes for the same object or leave empty for separate objects. |
| apply_mask |  | true | Apply the mask on the image. |
| output_format | jpeg, png, webp | png | The format of the generated image. |

## What can you do with SAM 3 Image in OpenClaw?

Prepare object selections from incoming images for recurring review work. Keep the source images and mask previews together so a person can check each batch.

### Product cutout preparation

Select a photographed product with a description such as 'table lamp.' Pass the mask to an editor for background removal or a new composition.

### Selective photo editing

Mask a jacket, chair, or other object before applying a color adjustment in an image editor. Review the selection before changing the pixels.

### Object count review

Find matching objects such as apples or chairs and use the masks to check a proposed count. Keep a marked preview so a person can spot missed or duplicate selections.

### Annotation drafts

Create initial object masks for an image dataset using short category descriptions. Have an annotator correct the selections before accepting the labels.

### Privacy redaction preparation

Locate faces, license plates, or screens before a separate pixelation step. Inspect the whole image for missed targets before publishing.

### Interior design selections

Isolate furniture in a room photo to prepare a mood board or editing selection. Use descriptions such as 'armchair' or 'floor lamp' for each pass.

### Wildlife image annotation

Draft animal masks in camera-trap stills with broad descriptions such as 'bird' or 'deer.' Treat exact species identification as a separate review task.

## Overview

- [fal](https://spiral.ai/providers/fal)
- [All OpenClaw guides](https://spiral.ai/integrations/openclaw)

## SAM 3 Image in other agents

- [Amp](https://spiral.ai/integrations/amp/sam-3-image)
- [ChatGPT](https://spiral.ai/integrations/chatgpt/sam-3-image)
- [Claude](https://spiral.ai/integrations/claude-desktop/sam-3-image)
- [Claude Code](https://spiral.ai/integrations/claude-code/sam-3-image)
- [Cline](https://spiral.ai/integrations/cline/sam-3-image)
- [Codex](https://spiral.ai/integrations/codex/sam-3-image)
- [Cursor](https://spiral.ai/integrations/cursor/sam-3-image)
- [Devin](https://spiral.ai/integrations/devin/sam-3-image)
- [dots](https://spiral.ai/integrations/dots/sam-3-image)
- [Factory Droid](https://spiral.ai/integrations/factory-droid/sam-3-image)
- [Gemini](https://spiral.ai/integrations/gemini/sam-3-image)
- [Gemini CLI](https://spiral.ai/integrations/gemini-cli/sam-3-image)
- [Google Antigravity](https://spiral.ai/integrations/antigravity/sam-3-image)
- [goose](https://spiral.ai/integrations/goose/sam-3-image)
- [Grok](https://spiral.ai/integrations/grok/sam-3-image)
- [Grok Build](https://spiral.ai/integrations/grok-build/sam-3-image)
- [Hermes](https://spiral.ai/integrations/hermes/sam-3-image)
- [Instinct](https://spiral.ai/integrations/instinct/sam-3-image)
- [Manus](https://spiral.ai/integrations/manus/sam-3-image)
- [Muse](https://spiral.ai/integrations/muse/sam-3-image)
- [Muse Code](https://spiral.ai/integrations/muse-code/sam-3-image)
- [n8n](https://spiral.ai/integrations/n8n/sam-3-image)
- [Notion](https://spiral.ai/integrations/notion/sam-3-image)
- [opencode](https://spiral.ai/integrations/opencode/sam-3-image)
- [Perplexity](https://spiral.ai/integrations/perplexity/sam-3-image)
- [Raycast](https://spiral.ai/integrations/raycast/sam-3-image)
- [VS Code](https://spiral.ai/integrations/vscode/sam-3-image)
- [Windsurf](https://spiral.ai/integrations/windsurf/sam-3-image)
- [Zapier](https://spiral.ai/integrations/zapier/sam-3-image)

## More OpenClaw guides

- [RunDiffusion Photo Flux](https://spiral.ai/integrations/openclaw/rundiffusion-photo-flux)
- [SAM 3.1 Image](https://spiral.ai/integrations/openclaw/sam-3-1-image)
- [Rembg Enhance](https://spiral.ai/integrations/openclaw/rembg-enhance)
- [SAM 3.1 Image to RLE](https://spiral.ai/integrations/openclaw/sam-3-1-image-to-rle)
- [Recraft Vectorize](https://spiral.ai/integrations/openclaw/recraft-vectorize)
- [Sana](https://spiral.ai/integrations/openclaw/sana)

## Questions about SAM 3 Image in OpenClaw

**Can OpenClaw use SAM 3 Image?**

Yes. Add Spiral to OpenClaw as an MCP server, and OpenClaw can find and run SAM 3 Image like any built-in tool.

**How much does SAM 3 Image cost in OpenClaw?**

$0.0065 per run at the default inputs. You pay from your Spiral balance as you go, with no subscription, and a failed run is not charged. Your OpenClaw plan is separate.

**Do I need a fal account to use SAM 3 Image in OpenClaw?**

No. Spiral runs SAM 3 Image on its own fal account. You sign in to Spiral once and never paste a key.

**How long does it take to set up SAM 3 Image in OpenClaw?**

About two minutes: add the server, then sign in, then reload and check.

**Does OpenClaw wait while SAM 3 Image runs?**

SAM 3 Image runs in the background. OpenClaw starts the run, checks on it and gives you the image when it's ready.

**What does SAM 3 produce?**

SAM 3 produces segmentation masks that mark which pixels belong to matching objects in an existing image. Those selections can feed a separate editing, counting, or annotation step.

**How should I phrase a SAM 3 prompt?**

Use a short object description such as 'red apple' or 'striped cat.' Keep it to the object and one or two visible attributes.

**Can SAM 3 select several matching objects?**

Yes. SAM 3 is designed to find all instances matching a concept, such as the chairs in a room. Review the selections before treating them as a complete count.

**Can SAM 3 follow a detailed instruction about an object's position?**

Long descriptions such as 'the second to last book from the right on the top shelf' exceed its intended text prompting capabilities. Use a simpler description and visual guidance to identify the target.

**Where does SAM 3 need extra review?**

Specialized categories in unfamiliar imagery need particular care. Meta reports difficulty with fine-grained concepts requiring domain knowledge, so review scientific and medical selections closely.

