pinchtab Pass

Control a headless or headed Chrome browser via Pinchtab's HTTP API. Use for web automation, scraping, form filling, navigation, and multi-tab workflows. Pinchtab exposes the accessibility tree as flat JSON with stable refs — optimized for AI agents (low token cost, fast). Use when the task involves: browsing websites, filling forms, clicking buttons, extracting page text, taking screenshots, or any browser-based automation. Requires a running Pinchtab instance (Go binary).

77out of 100
1.0k
stars
27
downloads
84
views

// Install Skill

Install Skill

Skills are third-party code from public GitHub repositories. SkillHub scans for known malicious patterns but cannot guarantee safety. Review the source code before installing.

Install globally (user-level):

npx skillhub install openclaw/skills/pinchtab

Install in current project:

npx skillhub install openclaw/skills/pinchtab --project

Suggested path: ~/.claude/skills/pinchtab/

AI Review

77
out of 100
Instruction Quality82
Description Precision68
Usefulness79
Technical Soundness75

Scored 77 for comprehensive browser automation with agent-optimized accessibility tree and genuine multi-agent coordination via tab locking. Excellent 6-file progressive disclosure. Novel TRUST.md security transparency doc. Deduction for missing negative triggers in description and Go binary requirement not cross-platform.

betamoderatedevelopersai-agentsweb-scrapingbrowser-automationform-fillingmulti-agent-coordination
Reviewed by claude-code on 4/14/2026

SKILL.md Content

---
name: pinchtab
description: >
  Control a headless or headed Chrome browser via Pinchtab's HTTP API. Use for web automation,
  scraping, form filling, navigation, and multi-tab workflows. Pinchtab exposes the accessibility
  tree as flat JSON with stable refs — optimized for AI agents (low token cost, fast).
  Use when the task involves: browsing websites, filling forms, clicking buttons, extracting
  page text, taking screenshots, or any browser-based automation. Requires a running Pinchtab
  instance (Go binary).
homepage: https://pinchtab.com
metadata:
  openclaw:
    emoji: "🦀"
    requires:
      bins: ["pinchtab"]
---

# Pinchtab

Fast, lightweight browser control for AI agents via HTTP + accessibility tree.

## Setup

Ensure Pinchtab is running:

```bash
# Headless (default for automation)
BRIDGE_HEADLESS=true pinchtab &

# With UI (debugging)
pinchtab &
```

Default port: `9867`. Override with `BRIDGE_PORT=9868`.
Auth: set `BRIDGE_TOKEN=<secret>` and pass `Authorization: Bearer <secret>`.

Base URL for all examples: `http://localhost:9867`

## Core Workflow

The typical agent loop:

1. **Navigate** to a URL
2. **Snapshot** the accessibility tree (get refs)
3. **Act** on refs (click, type, press)
4. **Snapshot** again to see results

Refs (e.g. `e0`, `e5`, `e12`) are cached per tab after each snapshot — no need to re-snapshot before every action unless the page changed significantly.

## API Reference

### Navigate

```bash
curl -X POST http://localhost:9867/navigate \
  -H 'Content-Type: application/json' \
  -d '{"url": "https://example.com"}'
```

### Snapshot (accessibility tree)

```bash
# Full tree
curl http://localhost:9867/snapshot

# Interactive elements only (buttons, links, inputs) — much smaller
curl "http://localhost:9867/snapshot?filter=interactive"

# Limit depth
curl "http://localhost:9867/snapshot?depth=5"

# Smart diff — only changes since last snapshot (massive token savings)
curl "http://localhost:9867/snapshot?diff=true"

# Text format — indented tree, ~40-60% fewer tokens than JSON
curl "http://localhost:9867/snapshot?format=text"
```

Returns flat JSON array of nodes with `ref`, `role`, `name`, `depth`, `value`, `nodeId`.

**Token optimization**: Use `?filter=interactive` for action-oriented tasks (~75% fewer tokens). Use `?diff=true` on multi-step workflows to see only what changed. Use `?format=text` for cheapest structured output. Use full snapshot only when you need complete page understanding.

### Act on elements

```bash
# Click by ref
curl -X POST http://localhost:9867/action \
  -H 'Content-Type: application/json' \
  -d '{"kind": "click", "ref": "e5"}'

# Type into focused element (click first, then type)
curl -X POST http://localhost:9867/action \
  -H 'Content-Type: application/json' \
  -d '{"kind": "click", "ref": "e12"}'
curl -X POST http://localhost:9867/action \
  -H 'Content-Type: application/json' \
  -d '{"kind": "type", "ref": "e12", "text": "hello world"}'

# Press a key
curl -X POST http://localhost:9867/action \
  -H 'Content-Type: application/json' \
  -d '{"kind": "press", "key": "Enter"}'

# Focus an element
curl -X POST http://localhost:9867/action \
  -H 'Content-Type: application/json' \
  -d '{"kind": "focus", "ref": "e3"}'

# Fill (set value directly, no keystrokes)
curl -X POST http://localhost:9867/action \
  -H 'Content-Type: application/json' \
  -d '{"kind": "fill", "selector": "#email", "text": "[email protected]"}'

# Hover (trigger dropdowns/tooltips)
curl -X POST http://localhost:9867/action \
  -H 'Content-Type: application/json' \
  -d '{"kind": "hover", "ref": "e8"}'

# Select dropdown option (by value or visible text)
curl -X POST http://localhost:9867/action \
  -H 'Content-Type: application/json' \
  -d '{"kind": "select", "ref": "e10", "value": "option2"}'

# Scroll to element
curl -X POST http://localhost:9867/action \
  -H 'Content-Type: application/json' \
  -d '{"kind": "scroll", "ref": "e20"}'

# Scroll by pixels (infinite scroll pages)
curl -X POST http://localhost:9867/action \
  -H 'Content-Type: application/json' \
  -d '{"kind": "scroll", "scrollY": 800}'

# Click and wait for navigation (link clicks)
curl -X POST http://localhost:9867/action \
  -H 'Content-Type: application/json' \
  -d '{"kind": "click", "ref": "e5", "waitNav": true}'
```

### Extract text

```bash
# Readability mode (default) — strips nav/footer/ads, keeps article/main content
curl http://localhost:9867/text

# Raw innerText (old behavior)
curl "http://localhost:9867/text?mode=raw"
```

Returns `{url, title, text}`. Cheapest option (~1K tokens for most pages).

### Screenshot

```bash
# Raw JPEG bytes
curl "http://localhost:9867/screenshot?raw=true" -o screenshot.jpg

# With quality setting (default 80)
curl "http://localhost:9867/screenshot?raw=true&quality=50" -o screenshot.jpg
```

### Evaluate JavaScript

```bash
curl -X POST http://localhost:9867/evaluate \
  -H 'Content-Type: application/json' \
  -d '{"expression": "document.title"}'
```

### Tab management

```bash
# List tabs
curl http://localhost:9867/tabs

# Open new tab
curl -X POST http://localhost:9867/tab \
  -H 'Content-Type: application/json' \
  -d '{"action": "new", "url": "https://example.com"}'

# Close tab
curl -X POST http://localhost:9867/tab \
  -H 'Content-Type: application/json' \
  -d '{"action": "close", "tabId": "TARGET_ID"}'
```

Multi-tab: pass `?tabId=TARGET_ID` to snapshot/screenshot/text, or `"tabId"` in POST body.

### Health check

```bash
curl http://localhost:9867/health
```

## Token Cost Guide

| Method | Typical tokens | When to use |
|---|---|---|
| `/text` | ~800 | Reading page content |
| `/snapshot?filter=interactive` | ~3,600 | Finding buttons/links to click |
| `/snapshot?diff=true` | varies | Multi-step workflows (only changes) |
| `/snapshot?format=text` | ~40-60% less | Structured tree, cheaper than JSON |
| `/snapshot` | ~10,500 | Full page understanding |
| `/screenshot` | ~2K (vision) | Visual verification |

**Strategy**: Start with `/snapshot?filter=interactive`. Use `?diff=true` on subsequent snapshots in multi-step tasks. Use `/text` when you only need the readable content. Use `?format=text` to cut token costs further. Use full `/snapshot` only for complete page understanding.

## Environment Variables

| Var | Default | Description |
|---|---|---|
| `BRIDGE_PORT` | `9867` | HTTP port |
| `BRIDGE_HEADLESS` | `false` | Run Chrome headless |
| `BRIDGE_TOKEN` | (none) | Bearer auth token |
| `BRIDGE_PROFILE` | `~/.pinchtab/chrome-profile` | Chrome profile dir |
| `BRIDGE_STATE_DIR` | `~/.pinchtab` | State/session storage |
| `BRIDGE_NO_RESTORE` | `false` | Skip tab restore on startup |
| `CDP_URL` | (none) | Connect to existing Chrome DevTools |

## Tips

- Refs are stable between snapshot and actions — no need to re-snapshot before clicking
- After navigation or major page changes, take a new snapshot to get fresh refs
- Use `filter=interactive` by default, fall back to full snapshot when needed
- Pinchtab persists sessions — tabs survive restarts (disable with `BRIDGE_NO_RESTORE=true`)
- Chrome profile is persistent — cookies/logins carry over between runs

License

Declared license: MIT

MIT License

Copyright (c) 2026 openclaw

Permission is hereby granted, free of charge, to any person obtaining a copy
of this software and associated documentation files (the "Software"), to deal
in the Software without restriction, including without limitation the rights
to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
copies of the Software, and to permit persons to whom the Software is
furnished to do so, subject to the following conditions:

The above copyright notice and this permission notice shall be included in all
copies or substantial portions of the Software.

THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
SOFTWARE.

View the license in the source repositorythe version published there is authoritative.