firecrawl/opencode-firecrawl

最近记录的来源活动: 2026年10月1日 15:09

SkillsMP 收录数据更新: 2026年11月9日 14:22

已收集 skills: 6

GitHub 星标: 104
GitHub Forks: 6

用 AI 助手安装

复制这段 Prompt，发给你正在使用的 AI 助手。

Follow https://skillsmp.com/skill-install/prompt.md to install Agent Skills from https://github.com/firecrawl/opencode-firecrawl.

仓库 README

仓库 README 来源：README.md

查看 README 源文件

展开完整 README

# @firecrawl/opencode [OpenCode](https://opencode.ai) plugin for [Firecrawl](https://firecrawl.dev). It gives your agent primary-source answers from the Firecrawl developer index, structured data from Firecrawl Alexandria providers, and web scraping, crawling, and search through the [Firecrawl CLI](https://github.com/firecrawl/cli). ## Installation The plugin supports OpenCode 2 and OpenCode 1 (1.18.29 or later) from the same package. ### OpenCode 2 ```bash opencode plugin add @firecrawl/opencode ``` Or add it to `opencode.jsonc` yourself: ```jsonc { "$schema": "https://opencode.ai/config.json", "plugins": ["@firecrawl/opencode"] } ``` ### OpenCode 1 Add it to `opencode.json`: ```json { "$schema": "https://opencode.ai/config.json", "plugin": ["@firecrawl/opencode"] } ``` OpenCode installs the package on the next start. The developer search tool works right away, with no API key and no CLI. Without a key, requests from some networks (VPNs, proxies, cloud hosts) are refused; connect Firecrawl to search from there. For the web skills (scrape, crawl, map, search, agent), install the Firecrawl CLI: ```bash npx -y firecrawl-cli@latest init -y --browser ``` ## Authentication On OpenCode 2, run `/connect` in the TUI (or `opencode auth login`), pick Firecrawl, and choose **Sign in with Firecrawl**. OpenCode opens Firecrawl in your browser; approve access and you are connected, with no API key to copy. OpenCode keeps the sign-in fresh on its own. Pasting an API key from [firecrawl.dev](https://firecrawl.dev) works too, as does `FIRECRAWL_API_KEY` in the environment. Connecting or disconnecting takes effect without a restart. The browser sign-in is used by this plugin's tools and the Firecrawl CLI. OpenCode's built-in Firecrawl web search reads only an API key. On OpenCode 1, export an API key before starting OpenCode: ```bash export FIRECRAWL_API_KEY=fc-your-api-key ``` Once connected, the plugin: - registers the `firecrawl_alexandria` tool - passes the credential to shell commands, so the CLI is authenticated - raises the rate limit on developer search `firecrawl login --browser` authenticates the CLI alone. ## Tools ### `firecrawl_developer_search` Searches an index of GitHub issues, merged pull requests, READMEs, and library documentation, and returns the **matched passages** as markdown rather than a list of links. Use it for how a library behaves, what an error means, whether a bug was fixed, or what an API contract guarantees. ``` Why does my Playwright script hang on page.goto with a service worker registered? ``` It takes a `query` plus optional `types` (`doc`, `issue`, `pull_request`, `readme`), `repos` (`owner/name`), `sources`, `k`, and `passages`. No API key needed. ### `firecrawl_alexandria` Finds and runs [Alexandria](https://www.firecrawl.dev/alexandria) data providers: official APIs, licensed publishers, and Firecrawl indexes that return structured records instead of a web page. ``` What has the unemployment rate in California been over the last two years? ``` The agent passes `query` to discover matching capabilities with their input contract and price (free), then `provider`, `capability`, and `options` to run one. A run spends Firecrawl credits at the listed price. The tool only appears once a key is connected. Runs go through the `firecrawl_alexandria` permission, matched on `provider/capability`, and are allowed unless you configure it. Discovery never asks. On OpenCode 2 (2.0.29 or later), to confirm each run before it spends: ```jsonc { "permissions": [{ "action": "firecrawl_alexandria", "resource": "*", "effect": "ask" }] } ``` Answering "always" remembers that capability. Narrow `resource` to a provider (`"bls-gov/*"`) to ask for some runs only, and use `"effect": "deny"` to block them. Earlier OpenCode 2 releases cannot ask from a plugin tool. On OpenCode 1: ```json { "permission": { "firecrawl_alexandria": "ask" } } ``` ## Skills The plugin bundles the Firecrawl CLI skills (search, scrape, crawl, map, interact, download, parse, agent, monitor, the developer and research indexes, and Alexandria). They teach the agent when each one fits and how to keep output in a `.firecrawl/` directory instead of the context window. ## Links - [Firecrawl CLI documentation](https://docs.firecrawl.dev/sdks/cli) - [Firecrawl CLI on GitHub](https://github.com/firecrawl/cli) - [OpenCode plugin docs](https://opencode.ai/v2/docs/plugins) ## License ISC

正在加载仓库 README…

这个仓库中的 skills

职业分类待补全

skill

职业分类

描述

更新

firecrawl-agent

未分类

Autonomously navigate websites and extract structured data across pages. Use when the task requires navigation or no suitable ready-made workflow or data provider covers it.

原文语言：英语

2026-10-01

firecrawl-alexandria

未分类

Find a direct path to structured data through ready-made workflows, data APIs, and indexes. Follow the search skill to discover and inspect tools, then the scrape skill to execute them.

原文语言：英语

2026-10-01

firecrawl-developer-index

未分类

Search an index of public repositories, GitHub issues, merged pull requests, repository READMEs, and curated documentation sites. Use when a programming question needs external documentation or upstream evidence.

原文语言：英语

2026-10-01

firecrawl-scrape

未分类

Read a known webpage or execute a discovered workflow or data-provider capability. Use for page content or structured results once the URL or tool is selected.

原文语言：英语

2026-10-01

firecrawl-search

未分类

Find web sources with query-relevant page excerpts and optional full-page content, and discover workflows, data APIs, and indexes. Use for web research or finding structured records, listings, transcripts, and datasets. Supports semantic tool discovery,…

原文语言：英语

2026-10-01

firecrawl

未分类

Any live-web task via the Firecrawl CLI — including ordinary web research: searching the web, reading or extracting pages, gathering sources, discovering site URLs, bulk extraction, downloading a site, change alerts, or pages needing clicks/login — web only;…

原文语言：英语

2026-10-01

已展示 6 / 6 个已收集 Skill。