scrape

An agent skill by garrytan, from garrytan/gstack. Tags: browser, data, scraping, web.

What it does

Pull data from a web page through the Aside browser — your real, already signed-in sessions. (gstack)

Install

With the skills CLI, which installs into Claude Code, Codex, Cursor and other agents:

npx skills add garrytan/gstack --skill scrape

Or copy the skill folder into Claude Code's skills directory by hand (~/.claude/skills for every project, or .claude/skills inside one):

git clone --depth 1 https://github.com/garrytan/gstack
cp -r gstack/scrape ~/.claude/skills/scrape

Safety box score

Grade D, elevated risk: 51 of 100 risk points, rated 2026-10-16. The grade rates what the skill and its scripts can reach on the machine of whoever installs it, across eight categories from shell execution to secrets access. How the score works.

Scrape is a read-only browser-scraping skill, but it runs local gstack helper scripts, can use your signed-in browser sessions or imported cookies, and writes telemetry and learnings into home-directory paths. The main risks are local code execution, exposure of authenticated account data, and a prompt-injection surface from runtime tool output and page content.

Category levels, from 0 (none) to 3 (high)
Category	Level	Why
Code & shell execution	2 (moderate)	It invokes local helper programs like `~/.claude/skills/gstack/bin/gstack-skill-start`, `gstack-skill-end`, `gstack-learnings-log`, and may run `cd <SKILL_DIR> && ./setup`.
Network access	2 (moderate)	It drives a browser to remote URLs and can send data through remote browser services such as `aside exec`; it also logs telemetry to analytics paths.
Filesystem reach	2 (moderate)	It writes outside the working tree into `.gstack/tmp`, `.git/info/exclude`, and `$GSTACK_STATE_ROOT/analytics/`, and depends on home-directory paths under `~/.claude/skills/`.
Secrets & credentials	2 (moderate)	It is explicitly meant to use the user's "real, already signed-in sessions" and the fallback path says `setup-browser-cookies (imports real-browser cookies)`.
Privilege & persistence	0 (none)	There is no sudo, service installation, shell startup editing, or other privilege escalation in the skill text.
Instruction hijack surface	1 (low)	It ingests runtime `GSTACK_INSTRUCTION_BEGIN` blocks from tool output and page content is treated as data, which creates a small instruction-injection surface even though the skill warns against following page text.
Irreversible actions	2 (moderate)	It can click, fill, and on `LOCAL` targets the browser rules allow mutating actions like submit/create/delete to proceed, though non-local mutations need user confirmation.
Transparency	2 (moderate)	Important behavior is delegated to opaque files and binaries not included here, such as `~/.claude/skills/gstack/bin/*` and `./setup`.

Before installing

Inspect `~/.claude/skills/gstack/bin/gstack-skill-start`, `gstack-skill-end`, and `gstack-learnings-log` if they are present.
Confirm you are comfortable with the skill using your logged-in browser sessions or imported cookies.
Try it first on a non-sensitive page or disposable account.
Check whether writes to `.gstack/tmp` and `$GSTACK_STATE_ROOT/analytics/` are acceptable on your machine.

Rated 2026-10-16 by an automated review of 1 file from the skill's repository.

Source

Repository: garrytan/gstack (all skills from this repository)
Path: scrape/SKILL.md
Branch: main
Updated: 2026-10-05

competitor-profiling — When the user wants to research, profile, or analyze competitors from their URLs.
programmatic-seo — When the user wants to create SEO-driven pages at scale using templates and data.
seo-firecrawl — Full-site crawling, scraping, and site mapping via Firecrawl MCP.
hackernews-frontpage — Scrape the Hacker News front page (titles, points, comment counts).
open-gstack-browser — Launch GStack Browser — AI-controlled Chromium with the sidebar extension baked in.
setup-browser-cookies — Import cookies from your real Chromium browser into the headless browse session. (gstack)