parallel-web-extract Skill for Claude: What It Does and How to Install It

Makes Claude pull the full text of web pages and PDFs with parallel-cli and return it unchanged.

Installs: 13.8K; on skills.sh
Stars: 91; on GitHub
Updated: Oct 2026; last repo commit

This is Parallel's own CLI-backed extraction skill. Claude passes your URLs to parallel-cli extract, saves the result as JSON under /tmp, and returns the content verbatim with the page title and link. SKILL.md says to prefer the Parallel Search MCP web_fetch tool for ordinary pages when it is available.

It needs parallel-cli installed, signed in and funded on your Parallel account. The folder holds only SKILL.md and bundles no scripts.

Good for

Anyone who needs the actual text of articles, docs or PDFs inside the chat
Developers who want saved JSON and CLI-only options such as full-content caps
Users of JavaScript-heavy sites that plain fetch tools struggle with

Install parallel-web-extract

Claude Code

Copy this prompt.

Install the parallel-web-extract skill in Claude Code. Run this command:

/plugin marketplace add parallel-web/parallel-agent-skills
/plugin install parallel

Then check that it loaded and tell me how to use it.

2
Paste it into Claude Code. It shows the change and asks before running it.

Pick one route. The plugin installs all 15 skills (counted from the repo), a hosted Parallel Search MCP server (parallel-search, anonymous) and a Bash-only agent called parallel-subagent. Restart Claude Code after installing it. This skill needs parallel-cli installed, signed in and funded; the second method in each tab lists the install commands from the repo's setup skill, and /parallel:parallel-cli-setup runs them for you. In Codex, use $ instead of / for skill names. This skill names the agent parallel:parallel-subagent in its frontmatter. That agent file (agents/parallel-subagent.md) ships with the plugin, not inside this skill folder, so the plugin is the route listed first.

From the project README. Source.

What's in the skill

When it loads

CLI-backed URL extraction. Use when the user explicitly invokes this skill, needs CLI-only controls or saved JSON output, or no Parallel web_fetch MCP tool is available. When the bundled Parallel Search MCP is available, prefer its web_fetch tool for ordinary webpages, articles, PDFs, and JavaScript-heavy sites.

Sections in SKILL.md

Command
Handling failed extractions
Response format
Setup

Key facts

Author: Parallel Web Systems (vendor's own repo)
Repo: parallel-web/parallel-agent-skills
License: MIT (LICENSE file; README and plugin.json agree)
Installs on skills.sh: 13,764 (Nov 2, 2026)
GitHub stars: 91 for the whole repo (Nov 2, 2026)
Last repo commit: Oct 28, 2026
Audits on skills.sh: Gen Agent Trust Hub: pass, Socket: pass, Snyk: warn
What it includes: 1 file: SKILL.md. It bundles no scripts
Part of: Plugin parallel: 15 skills, a hosted MCP server and 1 agent
Needs: parallel-cli, internet access, a funded Parallel account
Runs in: A forked helper agent, parallel:parallel-subagent

What it includes

Fetches up to 20 URLs per call and saves the content to a JSON file
Each URL is its own quoted argument. Output goes to /tmp/<name>.json with -o.
Try: “Extract the content of https://docs.parallel.ai/integrations/cli”
Runs code · parallel-cli extract
Focus the excerpts on a goal or on keywords
Runs code · --objective and -q
Returns the whole page body, optionally capped with --full-content-max-chars
Use it for long articles and PDFs. Even full content may be capped upstream, so the skill must not promise completeness.
Runs code · --full-content
Strips excerpts when you only want full content
Runs code · --no-excerpts
Reports each failed URL and its reason while keeping the successes
Empty results are not a successful extraction, and the skill must not make content up.
Failure handling
Returns page content as written, minus menus, footers and ads
SKILL.md says never to paraphrase or summarize extracted content.
Verbatim output format
If parallel-cli is missing, runs /parallel:parallel-cli-setup
Runs code · Setup fallback

From the official docs

Permissions

Medium risk

What it can do

Read your data
Create and change things
Delete things
Read-only mode

Watch out for

Your queries are sent to Parallel
The README says task inputs go to Parallel's API under its Customer Terms and Privacy Policy, and tells you not to send secrets or sensitive local content. Review what Claude puts in a request. github.com
It runs against your funded Parallel account
The README says CLI-backed skills are billed as ordinary API usage on your own Parallel account. Check current pricing on the vendor site before large runs. github.com
The setup path installs and signs in a CLI
The repo's setup skill installs parallel-cli (brew, uv, npm or pipx), starts a device sign-in, and points to `parallel-cli balance add` for top-ups. The search and research skills say never to add funds without explicit confirmation. github.com
URLs you ask about are sent to Parallel
Parallel fetches the pages, so any URL Claude passes in, including private or internal ones, goes to Parallel. Share only public links. The skill passes up to 20 URLs per call. github.com

Technical details

Turning on read-only	There is no read-only switch. The skill runs parallel-cli commands that call Parallel's API under your account.
Sign-in and keys	parallel-cli signs in with a device OAuth flow you approve in a browser. The README says CLI-backed skills use credentials you already configured.
Scopes	Runs parallel-cli through Bash (allowed-tools: Bash(parallel-cli:*)) Writes extracted page content as JSON to a file under /tmp (or the path Claude picks)

Questions

What does the parallel-web-extract skill do?

It makes Claude pull the text of web pages, articles, PDFs or JavaScript-heavy sites with parallel-cli extract, then return it verbatim with the file path of the saved JSON.

How do I install parallel-web-extract?

Use the install box above. Install the Claude Code plugin and restart Claude Code, then install and sign in to parallel-cli. The skill names an agent that ships in the plugin.

Should I use this or the Parallel Search MCP web_fetch tool?

SKILL.md says that when the bundled Parallel Search MCP is available, Claude should prefer its web_fetch tool for ordinary webpages, articles, PDFs and JavaScript-heavy sites. Use this skill when you invoke it, need CLI-only controls or saved JSON, or no web_fetch tool exists.

How many URLs can I extract at once?

Up to 20 per call, each as a separate quoted argument. SKILL.md says not to join them into one string or use eval.

Does it summarize the page?

No. SKILL.md says to keep content verbatim and not paraphrase. Only obvious noise such as menus, footers and ads is stripped. Large pages are kept whole in the saved file with a short labelled preview in chat.

Is it free?

The CLI skill runs against your own funded Parallel account and is billed as ordinary API usage, per the README. The anonymous free tier covers the bundled MCP search and fetch tools, not this skill. See the vendor's pricing page.

Is it safe?

It runs parallel-cli in Bash, sends the URLs to Parallel, and writes a JSON file under /tmp. It spends credits on your account. Our advice: only pass public URLs.

Is it an official Parallel skill?

It is in Parallel Web Systems' own repo, and the repo's marketplace.json describes the plugin as the "Official Parallel Web Systems plugin for Claude Code".

Information

updated Nov 2, 2026

Repository: parallel-web/parallel-agent-skills
Folder: skills/parallel-web-extract

parallel-web/parallel-agent-skills skills.sh

Sources

Docs, issue trackers and discussions this page is built from.

Show all 6

parallel-agent-skills README · github.com
parallel-agent-skills marketplace.json · github.com
parallel plugin.json · github.com
parallel-cli-setup SKILL.md · github.com
parallel-web-extract SKILL.md · github.com
parallel-subagent.md · github.com