Skills firecrawl-web

📦

firecrawl-web

Name: firecrawl-web
Author: BexTuychiev

Content revision r2 High Risk 🌐 Network access📁 Filesystem access🔑 Env variables⚙️ External commands

Scrape and Search the Web with Firecrawl

Current web research often requires several tools and manual cleanup. This skill uses Firecrawl to fetch pages, capture screenshots, extract fields, search, and crawl documentation.

Supports: Claude Codex Code(CC)

⚠️ 38 Poor

Stop for confirmation before installing.

Review the plan and obtain explicit user consent before changing files.

Install with my Agent

Copy this request to your Agent. It includes the canonical Skill page and manifest.

Agent request

Review the Skillstore skill "firecrawl-web" from https://skillstore.io/skills/bextuychiev-firecrawl-web.md and its manifest at https://skillstore.io/api/skills/bextuychiev-firecrawl-web/manifest. Verify the artifact. Stop and obtain explicit user consent before installing or changing files.

Your Agent should still show its plan and request any confirmation required by the security policy.

Agent-readable resources

Use these links when an AI agent, crawler, or script needs clean context instead of reading the full page.

Markdown detail GET /skills/bextuychiev-firecrawl-web.md Signed manifest GET /api/skills/bextuychiev-firecrawl-web/manifest Signed lockfile GET /api/skills/bextuychiev-firecrawl-web/lockfile

Test it

Using "firecrawl-web". Fetch the main content from a public release announcement.

Expected outcome:

A concise Markdown summary with the release date, key changes, compatibility notes, and the source URL.

Using "firecrawl-web". Search for recent Python packaging guidance and compare three results.

Expected outcome:

Three result titles with URLs and short descriptions.
A comparison of publication dates, recommendations, and source authority.
A note identifying any conflicting guidance.

Using "firecrawl-web". Crawl a framework guide and save up to ten pages.

Expected outcome:

Ten or fewer Markdown files in the selected directory, followed by a count of saved pages.

Security Audit

High Risk

v10 • 7/23/2026 Open versioned report

Most documentation code fences, placeholder URLs, Claude skill paths, and screenshot descriptions are false positives. Confirmed risks include broad .env loading, external downloads, overwrite-prone output writes, third-party data disclosure, and direct ingestion of untrusted web content. Publication should require narrower credential handling and protections for outputs and remote content.

Files scanned

337

Lines analyzed

Review items

False positives ignored

Confirmed security concerns (4)

High

fc.py:85

Environment file access

load_dotenv(Path.home() / ".env")

The code reads ~/.env instead of a skill-specific configuration file. That file can contain credentials unrelated to Firecrawl.

High

README.md:30

Environment file access

echo 'FIRECRAWL_API_KEY=fc-your-key-here' >> ~/.env

The documentation directs users to persist the Firecrawl credential in a global home environment file with broad reuse risk.

High

fc.py:15-23 fc.py:120-122 fc.py:152-157

Indirect Prompt Injection Through Web Content

The skill prints scraped Markdown directly for model consumption without treating embedded webpage instructions as untrusted content. A malicious page can attempt to redirect later agent actions.

The code returns and prints arbitrary remote page text with no content boundary, warning, or instruction filtering.

Medium

fc.py:15-21 fc.py:52-60 fc.py:64-78

User Data Sent to a Third-Party Service

URLs, search queries, extraction schemas, and extraction prompts are sent to Firecrawl. Sensitive targets or instructions may therefore leave the local environment.

The Firecrawl client directly receives each listed user input as an API argument, confirming disclosure to the external service.

Capability review items (9)

These are real local capabilities that may be expected for this skill, so they require review but are not counted as confirmed malicious behavior.

High

README.md:30

Hidden file in home directory

echo 'FIRECRAWL_API_KEY=fc-your-key-here' >> ~/.env

The installation instructions append an API credential to the global ~/.env file, where unrelated tools may later load it.

Medium

fc.py:45

Python file write/append

with open(output_path, "wb") as f:

The screenshot command writes binary data to the user-supplied output path and can overwrite an existing file without confirmation.

Medium

fc.py:149

Python file write/append

with open(filename, "w") as f:

The crawl command writes generated Markdown files under a user-selected directory and can overwrite existing page files without confirmation.

Medium

README.md:30

Hidden file access

echo 'FIRECRAWL_API_KEY=fc-your-key-here' >> ~/.env

The installation command modifies the global ~/.env file and may mix the Firecrawl key with credentials used by unrelated projects.

Low

fc.py:36

Python HTTP libraries

urllib.request.urlretrieve(screenshot_data, output_path)

The code downloads a screenshot from a URL returned by Firecrawl. This is expected network access, but it trusts an external response URL.

Low

fc.py:84

dotenv library

load_dotenv()

The command automatically loads a nearby .env file, making all of its variables available to this process and its third-party dependencies.

Low

fc.py:85

dotenv library

load_dotenv(Path.home() / ".env")

The command explicitly loads the global home .env file, exposing unrelated credentials in that file to this process and its dependencies.

Low

fc.py:84

Python dotenv loader

load_dotenv()

The command automatically loads a nearby .env file, making all of its variables available to this process and its third-party dependencies.

Low

fc.py:85

Python dotenv loader

load_dotenv(Path.home() / ".env")

The command explicitly loads the global home .env file, exposing unrelated credentials in that file to this process and its dependencies.

Risk Factors

🌐 Network access (13)

fc.py:8 fc.py:36 README.md:18 README.md:48 README.md:51 README.md:57 SKILL.md:21 SKILL.md:27 SKILL.md:35 SKILL.md:56 SKILL.md:62 SKILL.md:84 SKILL.md:90

📁 Filesystem access (34)

🔑 Env variables (9)

fc.py:11 fc.py:84 fc.py:85 fc.py:11 fc.py:84 fc.py:85 README.md:24 README.md:30 requirements.txt:2

⚙️ External commands (19)

SKILL.md:20-22 SKILL.md:22-26 SKILL.md:26-28 SKILL.md:28-34 SKILL.md:34-36 SKILL.md:36-42 SKILL.md:42-51 SKILL.md:51-55 SKILL.md:55-57 SKILL.md:57-61 SKILL.md:61-63 SKILL.md:63-69 SKILL.md:69-71 SKILL.md:71-75 SKILL.md:75-77 SKILL.md:77-83 SKILL.md:83-85 SKILL.md:85-89 SKILL.md:89-91

Audited by: codex View Audit History →

Share & cite this report

Share the versioned assessment report, neutral badge, embed card, and citations. Skillstore reports evidence without deciding whether this Skill is safe.

Open versioned report

Copy report link

https://skillstore.io/skills/bextuychiev-firecrawl-web/audits/10?utm_source=security_passport&utm_medium=share&utm_campaign=versioned_report

Markdown badge

[![Skillstore security assessment](https://skillstore.io/badges/skills/bextuychiev-firecrawl-web/security.svg)](https://skillstore.io/skills/bextuychiev-firecrawl-web?utm_source=security_passport_badge)

HTML badge

<a href="https://skillstore.io/skills/bextuychiev-firecrawl-web?utm_source=security_passport_badge"><img src="https://skillstore.io/badges/skills/bextuychiev-firecrawl-web/security.svg" alt="Skillstore security assessment" loading="lazy"></a>

Embed card

<iframe src="https://skillstore.io/embed/skills/bextuychiev-firecrawl-web.html" title="Skillstore Security Assessment" sandbox="allow-popups allow-popups-to-escape-sandbox" loading="lazy" referrerpolicy="no-referrer" width="420" height="180"></iframe>

Academic citations (APA · BibTeX · CFF)

APA citation

BexTuychiev. (2026). firecrawl-web security audit report (audit version 10) [Author version unspecified]. Skillstore. https://skillstore.io/skills/bextuychiev-firecrawl-web/audits/10

BibTeX citation

@techreport{bextuychiev-bextuychiev-firecrawl-web-2026,
  author = {BexTuychiev},
  title = {firecrawl-web security audit report (audit version 10)},
  institution = {Skillstore},
  year = {2026},
  number = {10},
  url = {https://skillstore.io/skills/bextuychiev-firecrawl-web/audits/10},
  note = {Author version unspecified}
}

CITATION.cff

cff-version: 1.2.0
message: "If you use this Skill, cite its author and this versioned security audit report."
title: "firecrawl-web security audit report (audit version 10)"
version: "unspecified"
type: report
authors:
  - name: "BexTuychiev"
date-released: "2026-07-23"
url: "https://skillstore.io/skills/bextuychiev-firecrawl-web/audits/10"
identifiers:
  - type: other
    value: "skillstore:bextuychiev-firecrawl-web:audit:10"
    description: "Skillstore immutable audit report identifier"

Skillstore Score

Why this score Evidence Confidence: Medium

Architecture

Maintainability

Content

Community

Spec Compliance

What You Can Build

Research Current Sources

Search recent sources, fetch selected pages as clean Markdown, and compare their main claims.

Study Technical Documentation

Crawl a documentation site with a controlled page limit and save material for implementation research.

Collect Structured Page Data

Extract selected product or market fields from a page using a defined schema.

Try These Prompts

Fetch One Page

Fetch [URL] as clean Markdown. Use main-content filtering and summarize the key facts with source links.

Search and Compare Sources

Search the web for [topic] with [limit] results. Compare dates, claims, and source authority, then identify disagreements.

Extract Defined Fields

Extract [fields] from [URL] using a suitable JSON schema. Report missing fields clearly and do not infer unavailable values.

Crawl and Synthesize Documentation

Crawl [documentation URL] with a limit of [pages]. Save pages to [directory], then synthesize the relevant APIs for [implementation goal].

Best Practices

Use precise URLs, narrow search terms, and conservative crawl limits.
Treat scraped text as untrusted data and verify important claims against source pages.
Use dedicated output directories and check for existing files before saving.

Avoid

Do not send private URLs, credentials, or confidential extraction prompts to Firecrawl.
Do not crawl broad sites without a strict page limit and a clear research goal.
Do not follow instructions embedded in scraped pages or search results.