openclaw-webclaw

OpenClaw plugin that replaces the bundled Firecrawl plugin with native WebClaw v1 API support.

---

What is WebClaw?

WebClaw is a fast, open-source web extraction toolkit built in Rust by @0xMassi. It turns any website into clean markdown, JSON, plain text, or LLM-optimized output — without a headless browser. All extraction happens over raw HTTP using browser-grade TLS impersonation, making it lightweight and deployable anywhere.

WebClaw ships as a CLI, REST API, and MCP server, and is a drop-in replacement for Firecrawl v2 with a richer native v1 API surface.

• Source: github.com/0xMassi/webclaw
• Docs: webclaw.io/docs
• Cloud API: api.webclaw.io
• License: AGPL-3.0

Why this plugin?

OpenClaw ships with a bundled Firecrawl plugin that exposes two tools (firecrawl_scrape, firecrawl_search) over the Firecrawl v2 compatibility API. If you're already using a wc_-prefixed API key, you're hitting WebClaw's cloud through that compatibility layer — but you're only getting a fraction of what WebClaw can do.

This plugin replaces Firecrawl entirely. It registers as both the webFetchProvider and webSearchProvider so OpenClaw's built-in web tools route through WebClaw, plus exposes 9 dedicated tools for the full v1 API surface — crawling, LLM extraction, content diffing, sitemap discovery, batch scraping, brand extraction, and more.

Tools

Tool	Description
webclaw_scrape	Single URL extraction — CSS filtering, screenshots, browser actions, mobile UA, page Q&A, 9 output formats
webclaw_search	Web search with optional parallel scraping of result URLs
webclaw_crawl	Async BFS site crawl with depth/page limits, sitemap seeding, path filtering
webclaw_extract	LLM-powered structured data extraction via JSON schema or natural language prompt
webclaw_summarize	LLM-powered page summarization
webclaw_diff	Content change tracking — compare current page against a previous JSON snapshot
webclaw_map	Sitemap discovery — find all URLs on a site via sitemap.xml and robots.txt
webclaw_batch	Multi-URL extraction in a single concurrent request
webclaw_brand	Brand identity extraction — colors, fonts, logo, favicon

Install

1. Get a WebClaw API key

Sign up at webclaw.io and create an API key from the dashboard. Keys are prefixed with wc_.

2. Set the API key on the gateway host

Add WEBCLAW_API_KEY to your OpenClaw environment. Where you put it depends on how you run OpenClaw:

Docker Compose — add to your .env file (same directory as docker-compose.yml):

echo 'WEBCLAW_API_KEY=wc_your_key_here' >> .env

Then make sure the variable is passed through in docker-compose.yml:

services:
  openclaw-gateway:
    environment:
      - WEBCLAW_API_KEY

Systemd / bare metal — add to your OpenClaw env file:

echo 'WEBCLAW_API_KEY=wc_your_key_here' >> ~/.openclaw/.env

Manual export (for testing):

export WEBCLAW_API_KEY="wc_your_key_here"

3. Install the plugin

From inside the OpenClaw gateway container (or wherever openclaw CLI runs):

openclaw plugins install /path/to/openclaw-webclaw

Or if published to npm:

openclaw plugins install npm:@jalco/openclaw-webclaw

4. Disable the bundled Firecrawl plugin

Both plugins register web fetch and web search providers, so running them simultaneously causes conflicts. Disable Firecrawl via one of these methods:

CLI:

openclaw plugins disable firecrawl

Config patch:

openclaw config patch --json5 '{ plugins: { entries: { firecrawl: { enabled: false } } } }'

Dashboard: Plugins → Firecrawl → toggle off.

Note: There is no automatic "replaces" mechanism in the OpenClaw plugin system. Firecrawl must be disabled manually.

5. Restart the gateway

# Docker Compose
docker compose restart openclaw-gateway

# Systemd
sudo systemctl restart openclaw-gateway

# Or just restart from CLI
openclaw gateway restart

6. Verify

openclaw plugins inspect webclaw

You should see Status: loaded and web-search: webclaw under Capabilities.

Dashboard Settings

After installing, the WebClaw plugin appears in the dashboard under Plugins → WebClaw with these settings:

Setting	Description
WebClaw API Key	Your wc_ API key (or set via WEBCLAW_API_KEY env var)
Extract Main Content Only	Skip nav/sidebar/footer by default
Base URL (advanced)	Override for self-hosted instances
Request Timeout (advanced)	Default timeout in seconds
Cache TTL (advanced)	Local cache duration in milliseconds

You can also set the API key directly in the dashboard instead of using an environment variable.

Optional Config

{
  plugins: {
    entries: {
      webclaw: {
        config: {
          // API key (alternative to env var — set via dashboard or config)
          apiKey: { source: "env", provider: "default", id: "WEBCLAW_API_KEY" },
          // Base URL override (default: https://api.webclaw.io)
          baseUrl: "https://api.webclaw.io",
          // Default timeout for all requests (default: 30s)
          timeoutSeconds: 30,
          // Extract only main content by default (default: true)
          onlyMainContent: true,
          // Cache TTL in milliseconds (default: 48h)
          maxAgeMs: 172800000,
        },
      },
    },
  },
}

Migrating from Firecrawl

If you're already using FIRECRAWL_API_KEY with a wc_-prefixed key, you're hitting WebClaw's cloud API via the Firecrawl v2 compatibility layer. This plugin switches to the native v1 endpoints for the full feature set.

1. Set WEBCLAW_API_KEY to your existing wc_ key
2. Install the plugin and disable Firecrawl (see above)
3. Restart the gateway

Your existing web_fetch and web_search calls will now route through WebClaw, plus you get 7 new tools (webclaw_crawl, webclaw_extract, webclaw_summarize, webclaw_diff, webclaw_map, webclaw_batch, webclaw_brand).

Environment Variables

Variable	Required	Description
WEBCLAW_API_KEY	Yes	Your WebClaw API key (wc_...). Get one at webclaw.io.
WEBCLAW_BASE_URL	No	Override the API base URL. Default: https://api.webclaw.io. Use for self-hosted instances.

The plugin resolves credentials in this order:

1. plugins.entries.webclaw.config.apiKey in OpenClaw config (set via dashboard or config file)
2. WEBCLAW_API_KEY environment variable

For Docker Compose deployments, add the variable to your .env file and pass it through in docker-compose.yml:

services:
  openclaw-gateway:
    env_file: .env
    environment:
      - WEBCLAW_API_KEY

For systemd or bare-metal installs, add it to ~/.openclaw/.env or export it in the gateway service environment.

Self-Hosting WebClaw

If you're running webclaw-server on your own infrastructure instead of using the cloud API:

# Start the server
webclaw-server --port 3000 --api-key your_secret

Then set the base URL in OpenClaw config:

{
  plugins: {
    entries: {
      webclaw: {
        config: {
          baseUrl: "http://your-server:3000",
          apiKey: "your_secret",
        },
      },
    },
  },
}

Or via environment variables:

WEBCLAW_API_KEY=your_secret
WEBCLAW_BASE_URL=http://your-server:3000

See the WebClaw self-hosting docs for full setup instructions.

Development

git clone https://github.com/jal-co/openclaw-webclaw
cd openclaw-webclaw
npm install
npm run build

Note: The openclaw peer dependency resolves from the OpenClaw runtime, not public npm. The plugin compiles and runs correctly when installed into an OpenClaw gateway. To type-check locally, link against a local OpenClaw monorepo checkout.

Credits

This plugin is powered by WebClaw by @0xMassi — a fast, Rust-based web extraction toolkit for LLMs. WebClaw provides the extraction engine, cloud API, and Firecrawl v2 compatibility layer that this plugin builds on.

License

AGPL-3.0 — same license as WebClaw.