WebLoom — Open Infrastructure for Web Automation

№ 01

Install

WebLoom ships as an MCP server. The fastest way to install: download webloom-install.md and drop it into your AI chat (Claude Code, Cursor, Continue, Cline). The assistant clones the engine, creates the folders, and wires the MCP config for you.

⬇ Download webloom-install.md

Or wire it manually with this config:

~/.mcp.json

{
  "mcpServers": {
    "webloom": {
      "type": "stdio",
      "command": "python",
      "args": ["path/to/webloom/server.py"],
      "env": {
        "ANTHROPIC_API_KEY": "<for vision layer 2.5>"
      }
    }
  }
}

On next message you should see 62 tools appear under the webloom namespace.

On first run WebLoom creates a folder at ~/.webloom/:

~/.webloom/

├── playbook.json          # live learning (auto-built)
├── threads/               # installed profile packs
│   ├── kdp.amazon.com.thread.json
│   ├── draft2digital.com.thread.json
│   └── …
├── sessions.json          # chrome session configs
└── startup-check.json

№ 02

Architecture

Two layers, top to bottom:

Threads

profile packs

JSON knowledge about ONE site. Selectors, endpoints, quirks, click + fill strategies, anti-bot pacing. Merged into the live playbook at runtime; user observations override Thread defaults.

WebLoom

engine

62 tools covering click ladder, 4 upload strategies, React/Redux/AUI/Backbone awareness, vision Layer 2.5, network capture, recipe record/replay. Stays lean — complexity migrates to Threads.

№ 03

Thread schema

A Thread is a single JSON file with one required field — domain — and a fistful of optional fields the engine knows how to consume:

example.com.thread.json

{
  "domain": "kdp.amazon.com",
  "name": "KDP Auto-Publish Profile",
  "version": "1.0.0",
  "author": "username",
  "license": "proprietary | mit | cc-by | …",
  "framework": "amazon-aui",
  "default_strategy": "js",
  "notes": [
    "AjaxInput clears input.files after onchange — use xhr_upload",
    "Modal save needs A.declarative.fire('a:click', target)"
  ],
  "selectors": {
    "title_input": "#data-print-book-title",
    "categories_button": "#categories-modal-button"
  },
  "actions": {
    "submit_form": {
      "type": "aui_dispatch",
      "event": "a:click",
      "target": "#submit-button"
    }
  },
  "captured_endpoints": [
    { "type": "fetch", "url": "https://kdp.amazon.com/api/…" }
  ],
  "click_log": {
    "Choose categories": {
      "strategy": "cdp", "successes": 12, "failures": 1, "last_at": 1716000000
    }
  }
}

№ 04

Author a thread

The fastest path is to let WebLoom build a starter for you:

from any session

> Use webloom to visit https://example.com
> seed_from_tab(session=..., capture_seconds=4, save=true)

→ Generated Thread draft for example.com → ~/.webloom/threads/example.com.thread.json
  framework: react-17+  ·  anti-bot: normal  ·  ax elements: 23  ·  endpoints captured: 11

Then refine: edit notes, add quirks, document upload endpoints and submit actions. Once the Thread covers your workflow, export it as a portable file:

from any session

> export_thread(
    domain="example.com",
    name="Example Pro Profile",
    version="1.0.0",
    author="@you",
    license="proprietary"
  )

→ Wrote ~/.webloom/threads/example.com.thread.json
  12 keys captured · ready to share, install on another machine, or sell.

№ 05

Sell a thread

Upload your .thread.json to the Atelier. Set a price, choose a license. WebLoom holds 25% as a marketplace fee; 75% lands in your payout account.

Payments run through Polar (Merchant of Record). They handle global tax, chargebacks, refunds. You handle: nothing operational. Sites drift — buyers report it in one click, and the engine can self-repair a Thread in minutes. Your reputation and new sales stay intact.

Creator dashboard + Polar integration land in v0.2.

№ 06

Engine tools

62 tools, grouped:

Interaction

›click — actionability + CDP + JS dispatch + vision
›fill — React-aware native setter
›key_type — 3 modes (keystrokes, insertText, fast)
›key_press — Enter, Tab, Esc, Arrows, ...
›scroll_tab, screenshot, navigate

Upload

›upload_file — Strategy A/B/C/D
›xhr_upload — direct FormData to endpoint
›replay_xhr — replay captured calls

Detection

›scan_tab — full or ax-tree mode
›scan_tab_diff — incremental
›framework_detect — React/Redux/AUI/Backbone/Radix
›detect_anti_bot — Cloudflare/DataDome/Akamai
›detect_blocker — CAPTCHA/2FA prompts

State manipulation

›react_force_change — fiber-walk onChange
›react_inspect_store — 4-tier discovery
›redux_dispatch — direct action dispatch
›aui_dispatch — A.declarative.fire
›backbone_inspect

Mobile / touch

›touch_tap — CDP Input.dispatchTouchEvent

Recipes & Threads

›start_recording / end_recording
›replay_recipe
›seed_from_tab — author thread from current tab
›list_threads / install_thread / export_thread
›save_playbook / get_playbook / note

Auth & session

›auth_totp — pyotp
›pause_for_human — clean handoff
›export_profile / import_profile

Network & timing

›capture_network_start / stop
›get_captured_requests
›wait_for / wait_for_idle