← Back to Skills Marketplace
Web Researcher Mini
by
davidlin100-sketch
· GitHub ↗
· v1.0.1
330
Downloads
0
Stars
1
Active Installs
2
Versions
Install in OpenClaw
/install web-researcher-mini
Description
Firecrawl CLI for web scraping, crawling, and search. Scrape single pages or entire websites, map site URLs, and search the web with full content extraction....
Usage Guidance
This skill is an instruction-only adapter for the Firecrawl CLI and appears to do what it says (web scraping/search). However, the documentation expects a FIRECRAWL_API_KEY (and optional APIFY tokens) but the skill metadata does not declare these required credentials — so the agent will likely ask you for secrets at runtime. Before proceeding: (1) verify the firecrawl CLI package and its publisher on npm or the project's official website rather than trusting this skill bundle alone; (2) avoid pasting long-lived secrets into chat — prefer using the CLI's browser login or provide a limited-scope API key; (3) if you must set FIRECRAWL_API_KEY, consider using a session-specific or revocable key and export it only in your current shell (not persisted) until you confirm the tool's behavior; (4) be aware the instructions recommend adding the key to shell profiles and creating .firecrawl files — these are persistent actions you must perform yourself; (5) if you want stronger assurance, ask the skill author/source for the official upstream URL or package hash and inspect the CLI repository before installing.
Capability Analysis
Type: OpenClaw Skill
Name: web-researcher-mini
Version: 1.0.1
The bundle provides legitimate tools (firecrawl-cli and summarize) for web scraping and AI-driven summarization. The instructions are well-structured for an AI agent, promoting efficient resource management (e.g., using grep/head for large files) and standard authentication workflows. No evidence of malicious intent, data exfiltration, or harmful prompt injection was found; the use of environment variables and package managers (npm, brew) is consistent with the stated research purpose.
Capability Assessment
Purpose & Capability
The name/description (web scraping, crawling, search) align with the provided runtime instructions (firecrawl CLI commands for search, scrape, crawl, map). No unexpected capabilities (e.g., cloud admin, crypto) are requested.
Instruction Scope
The SKILL.md and install rules instruct the agent to perform authentication flows (firecrawl login --browser) and to ask the user for API keys, persist them via shell profile edits, and create/consume .firecrawl output files. That is expected for a CLI but grants the skill the explicit ability to prompt for and receive secrets from the user; the instructions also encourage storing keys in ~/.zshrc or ~/.bashrc. The skill tells agents to open browser-based auth without additional user confirmation and to request secrets if browser auth fails — this expands the agent's runtime scope to collecting user credentials.
Install Mechanism
There is no automated install spec or downloaded code in the skill bundle (instruction-only). The README suggests installing the CLI via npm, which is a standard third-party install path but is not executed by the skill itself.
Credentials
The skill documentation expects a FIRECRAWL_API_KEY (and mentions optional APIFY tokens), but the skill metadata lists no required environment variables or primary credential. This mismatch means the skill will ask for/expect secrets at runtime even though none are declared — a transparency gap. The number and purpose of the env vars mentioned in docs are proportional to a scraping service, but their omission from metadata is an inconsistency that increases risk for less-technical users.
Persistence & Privilege
The skill is not always-enabled and does not request elevated platform privileges. However, the install instructions explicitly recommend persisting API keys to shell startup files and creating a .firecrawl directory in working directories, which are persistent changes the user would make themselves. The skill does not itself perform these writes (instruction-only), but it guides the user to persistent configuration changes.
How to Use
- Make sure OpenClaw is installed (local or Docker)
- Run the install command in chat:
/install web-researcher-mini - After installation, invoke the skill by name or use
/web-researcher-mini - Provide required inputs per the skill's parameter spec and get structured output
Version History
v1.0.1
Initial release: Firecrawl CLI for advanced web scraping, crawling, and search.
- Introduced comprehensive `firecrawl` CLI commands for search, scrape, crawl, and map.
- Added installation and authentication instructions.
- Organized usage examples for each operation, including output management.
- Documented key CLI options for web search, selective scraping, site mapping, and content extraction.
- Removed generic web development docs; replaced with Firecrawl-specific skill documentation.
v1.0.0
Initial release of Firecrawl CLI skill for comprehensive web scraping, crawling, and search.
- Introduces search, scrape, crawl, and map commands with extensive options.
- Provides clean markdown and structured data optimized for LLMs.
- Supports installation, authentication, rate limit checks, and organization of results.
- Enables advanced operations: JavaScript rendering, content extraction, geo/location filters, time-based search, and credit tracking.
- Example commands and best practices included for each operation.
Metadata
Frequently Asked Questions
What is Web Researcher Mini?
Firecrawl CLI for web scraping, crawling, and search. Scrape single pages or entire websites, map site URLs, and search the web with full content extraction.... It is an AI Agent Skill for Claude Code / OpenClaw, with 330 downloads so far.
How do I install Web Researcher Mini?
Run "/install web-researcher-mini" in the OpenClaw or Claude Code chat to install it in one step — no extra setup required.
Is Web Researcher Mini free?
Yes, Web Researcher Mini is completely free (open-source). You can download, install and use it at no cost.
Which platforms does Web Researcher Mini support?
Web Researcher Mini is cross-platform and runs anywhere OpenClaw / Claude Code is available (cross-platform).
Who created Web Researcher Mini?
It is built and maintained by davidlin100-sketch (@weilianglin100-sketch); the current version is v1.0.1.
More Skills