AI Agent Hub
Back to plugins
🧰

dsh-firecrawl

Web Tools Updated 2026.08.17

Run the following command in DeepSeek Harness:

dsh plugin install zoahdev/dsh-firecrawl

Paste the following prompt into your AI chat to install this plugin:

Run dsh plugin install zoahdev/dsh-firecrawl in your terminal to add the plugin to DeepSeek Harness. The source repository is https://github.com/zoahdev/dsh-firecrawl .

About this plugin

LLMs are great at reasoning but struggle to read a live webpage. JavaScript-heavy pages dump raw HTML at the model, which is noisy, slow, and easy to truncate. dsh-firecrawl closes that gap by wrapping the Firecrawl Cloud API into native DeepSeek Harness tools, so the model receives clean, structured Markdown in a single turn.

It registers two tools. firecrawl_scrape takes a URL and lets Firecrawl handle server-side rendering and extraction, returning LLM-ready Markdown. firecrawl_search performs a web search and brings back the content of matching pages, removing the manual loop of fetching results one by one. Both call the cloud API over HTTPS, so no local browser or rendering engine is required.

If you are building RAG pipelines on DeepSeek Harness, doing real-time content summarization, or letting a model pull the latest documentation or competitor pages, this plugin flattens the path from web to model. With just a Firecrawl API key, it acts as a lightweight, transparent bridge that returns the full response without streaming or polling, making its behavior simple and predictable.

Use Cases

  • Let a model read the latest docs or tech blogs on demand and get clean Markdown
  • Pull external page content in real time as context for RAG pipelines
  • Search the web and summarize results within a single conversation

Best For

  • Developers building RAG or retrieval-augmented apps on DeepSeek Harness
  • Product or engineering teams that need LLMs to consume live web content
  • Users with a Firecrawl API key looking for a lightweight integration layer