From 26cc4d9166f48cd82803341e1837a1258045544c Mon Sep 17 00:00:00 2001 From: rramprakash Date: Tue, 9 Jun 2026 21:35:06 +0530 Subject: [PATCH 1/2] Docs: rewrite README for clarity and accuracy MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit - Fix stale org links (mendableai/* -> firecrawl/*) - Replace the partial/drifting CLI command table with a link to the canonical docs (docs.firecrawl.dev/cli) - List the actual shipped capabilities/skills (adds interact, agent, monitor, parse, download — previously missing or inaccurate) - Clarify install: add plugin -> install CLI -> sign in (browser, with API-key/env alternatives) - Add a Configuration table (incl. FIRECRAWL_NO_TELEMETRY opt-out) and tidy Resources/Support - Set license to MIT for consistency with the manifest and sibling plugins Co-Authored-By: Claude Opus 4.8 (1M context) --- README.md | 96 ++++++++++++++++++++++++++++++++++++++++++++++++------- 1 file changed, 84 insertions(+), 12 deletions(-) diff --git a/README.md b/README.md index 1a33420..b775052 100644 --- a/README.md +++ b/README.md @@ -1,24 +1,96 @@ -# Firecrawl Cursor Plugin +# Firecrawl for Cursor -Cursor plugin that gives AI agents full web access through the [Firecrawl CLI](https://github.com/mendableai/firecrawl-cli): search, scrape, map, crawl, browser automation, and AI-powered extraction. +Turn any website into clean, LLM-ready markdown or structured data — directly in Cursor. -## What's included +This plugin gives Cursor's agent full web access through the [Firecrawl CLI](https://github.com/firecrawl/cli): search, scrape, crawl, map, interact, monitor, and AI-powered extraction. Every operation includes JavaScript rendering, anti-bot handling, and proxy rotation. + +## Install + +### 1. Add the plugin -- **`skills/firecrawl/SKILL.md`** - Complete CLI reference covering search, scrape, map, crawl, browser, and agent commands with workflow patterns and examples -- **`rules/install.mdc`** - Installation and authentication handling +In Cursor (2.5+), run **`/add-plugin`** and choose **firecrawl**, or browse the [Cursor Marketplace](https://cursor.com/marketplace). -## Setup +### 2. Install the Firecrawl CLI + +The plugin's skills call the Firecrawl CLI, so install it globally: ```bash npm install -g firecrawl-cli +``` + +### 3. Sign in + +```bash firecrawl login --browser ``` +Prefer not to use the browser? Authenticate with a key instead: + +```bash +firecrawl login --api-key "fc-YOUR-API-KEY" +# or persist it as an environment variable +export FIRECRAWL_API_KEY=fc-YOUR-API-KEY +``` + +Get a free API key at [firecrawl.dev/app/api-keys](https://firecrawl.dev/app/api-keys). Verify everything's ready with `firecrawl --status`. + ## Capabilities -- **Search** - Web, image, and news search with optional full-page scraping of results -- **Scrape** - Single page content extraction to clean markdown, with JS rendering support -- **Map** - Discover all URLs on a site, with search filtering -- **Crawl** - Bulk content extraction from entire site sections -- **Browser** - Remote Chromium sessions for interactive pages (pagination, forms, modals) -- **Agent** - AI-powered autonomous web research and structured data extraction +Once installed, Cursor uses Firecrawl automatically for web tasks: + +- **Search** — web, news, and image search, with optional full-page scraping +- **Scrape** — clean markdown (or structured JSON) from any page, with JS rendering +- **Map** — discover every URL on a site +- **Crawl** — bulk-extract whole site sections +- **Interact** — click, fill forms, paginate, and log in on live pages +- **Agent** — AI-powered autonomous research and structured extraction +- **Monitor** — watch pages for meaningful changes, with webhook/email alerts +- **Parse** — convert local files (PDF, DOCX, XLSX, …) to markdown +- **Download** — save an entire site to local files + +## Usage + +Just ask Cursor naturally: + +``` +Search for "best practices for React testing" and summarize the key recommendations +Scrape https://docs.firecrawl.dev/introduction and pull out the main concepts +Map all the URLs under https://firecrawl.dev/blog +Monitor https://news.ycombinator.com and alert me when an AI story hits the top 10 +``` + +## What's included + +- **`skills/`** — 10 agent skills covering search, scrape, map, crawl, interact, agent, monitor, parse, download, plus a CLI workflow guide +- **`rules/install.mdc`** — sets up the Firecrawl CLI and handles authentication automatically + +## Command reference + +The skills cover the full command set. For every command and flag, see the [Firecrawl CLI docs](https://docs.firecrawl.dev/cli). + +## Configuration + +| Variable | Description | +|---|---| +| `FIRECRAWL_API_KEY` | Your API key (alternative to `firecrawl login`) | +| `FIRECRAWL_API_URL` | Custom endpoint for self-hosted instances | +| `FIRECRAWL_NO_TELEMETRY` | Set to `1` to disable anonymous CLI telemetry | + +**Self-hosted:** point the CLI at your instance with `export FIRECRAWL_API_URL=https://your-instance.com` (custom URLs skip API-key auth). + +## Resources + +- [Firecrawl CLI docs](https://docs.firecrawl.dev/cli) +- [Firecrawl CLI repository](https://github.com/firecrawl/cli) +- [API reference](https://docs.firecrawl.dev/api-reference) +- [Get an API key](https://firecrawl.dev/app/api-keys) + +## License + +MIT + +## Support + +- [Firecrawl Discord](https://discord.gg/gSmWdAkdwd) +- [GitHub Issues](https://github.com/firecrawl/firecrawl-cursor-plugin/issues) +- hello@firecrawl.dev From 3fafa74d4f3d101434e7f77599a4d99c02d13d21 Mon Sep 17 00:00:00 2001 From: rramprakash Date: Tue, 9 Jun 2026 21:38:35 +0530 Subject: [PATCH 2/2] =?UTF-8?q?Docs:=20correct=20Cursor=20install=20?= =?UTF-8?q?=E2=80=94=20use=20firecrawl=20CLI=20init=20(not=20/add-plugin)?= MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit /add-plugin and the Cursor plugin Marketplace aren't available to users yet. The supported install today is the Firecrawl CLI's init, which installs the skills into Cursor (.cursor/skills) and authenticates: npx -y firecrawl-cli@latest init --all --browser Co-Authored-By: Claude Opus 4.8 (1M context) --- README.md | 25 +++++++------------------ 1 file changed, 7 insertions(+), 18 deletions(-) diff --git a/README.md b/README.md index b775052..53f8746 100644 --- a/README.md +++ b/README.md @@ -6,33 +6,22 @@ This plugin gives Cursor's agent full web access through the [Firecrawl CLI](htt ## Install -### 1. Add the plugin - -In Cursor (2.5+), run **`/add-plugin`** and choose **firecrawl**, or browse the [Cursor Marketplace](https://cursor.com/marketplace). - -### 2. Install the Firecrawl CLI - -The plugin's skills call the Firecrawl CLI, so install it globally: +Set up the Firecrawl skills in Cursor and sign in with one command: ```bash -npm install -g firecrawl-cli +npx -y firecrawl-cli@latest init --all --browser ``` -### 3. Sign in +This installs the Firecrawl skills into Cursor (`.cursor/skills`), adds the install rules, and opens your browser to authenticate. **Restart Cursor afterward** so it picks up the new skills. -```bash -firecrawl login --browser -``` - -Prefer not to use the browser? Authenticate with a key instead: +Prefer to set it up manually? Install the CLI and sign in yourself: ```bash -firecrawl login --api-key "fc-YOUR-API-KEY" -# or persist it as an environment variable -export FIRECRAWL_API_KEY=fc-YOUR-API-KEY +npm install -g firecrawl-cli +firecrawl login --browser # or: firecrawl login --api-key "fc-YOUR-API-KEY" ``` -Get a free API key at [firecrawl.dev/app/api-keys](https://firecrawl.dev/app/api-keys). Verify everything's ready with `firecrawl --status`. +Get a free API key at [firecrawl.dev/app/api-keys](https://firecrawl.dev/app/api-keys). Verify with `firecrawl --status`. ## Capabilities