HighGrabber

Bulk-download from Hightail Spaces receive links. Give it a URL, a list of URLs, or a messy document with links buried in prose — HighGrabber finds them, logs you in once via your browser, and pulls every file to a destination you choose (auto-unzipping by default).

Works on macOS, Windows, and Linux.

---

Why

Hightail's web UI lets you download files one space at a time, serially, through the browser. If a friend shares fifty concert archives, clicking through fifty tabs and drag-extracting fifty ZIPs is a long afternoon. HighGrabber does it in one command.

It handles the two annoying parts of automating Hightail:

1. Login is gated by a WAF + CAPTCHA, so scripted HTTP login won't work. HighGrabber opens a real Chromium window the first time, you sign in once, and it reuses the session cookie on every run afterward.
2. Hightail rate-limits aggressive clients by returning HTTP 200 with an empty body. HighGrabber detects this, backs off exponentially, and resumes partial downloads via Range headers.

Install

Requires Python 3.10+.

With pipx (recommended)

pipx install highgrabber
highgrabber doctor        # downloads the browser used for login

Prefer to pin to a specific version? pipx install highgrabber==0.1.1.

With uv

uv tool install highgrabber
highgrabber doctor

From source (or a branch)

pipx install git+https://github.com/auxren/HighGrabber.git
highgrabber doctor

highgrabber doctor installs Playwright's Chromium into the right virtual environment and verifies your session is good. Run it once after install and any time you see a browser-related error.

Quickstart

# One URL
highgrabber https://spaces.hightail.com/receive/u3l28vnWZd -d ~/Downloads/show

# Many URLs from a file (prose is fine; it finds the links)
highgrabber message.txt -d ~/Downloads/shows

# Piped in
pbpaste | highgrabber -d ~/Downloads/shows

# Multiple inputs at once
highgrabber url1 url2 file.txt -d ~/Downloads/shows

The first run opens a Chromium window for you to log in to Hightail. Once the session is saved (~/.config/HighGrabber/storage_state.json on Linux, ~/Library/Application Support/HighGrabber/ on macOS, %APPDATA%\HighGrabber\ on Windows), subsequent runs are fully silent.

Usage

highgrabber [download] [INPUTS...] [options]
highgrabber login    [--email EMAIL] [--save-password]
highgrabber logout   [--forget-password --email EMAIL]

Inputs

Each input is one of:

• a full Hightail URL (https://spaces.hightail.com/receive/<slug> or https://spaces.hightail.com/space/<slug>)
• a path to any text file — HighGrabber extracts every Hightail link found anywhere in it, so forwarded emails, chat exports, and Notes pages all work
• - to read from stdin

Slugs are de-duplicated across all inputs in order.

Options (download)

flag	default	description
-d, --dest PATH	cwd	where to save files
-c, --concurrency N	2	parallel file downloads. Higher than 3 is usually rate-limited.
--email EMAIL	prompt	Hightail email (used for keychain lookup)
--save-password	off	save password to the system keychain on first login
--no-extract	off	keep ZIPs intact instead of auto-extracting
--delete-zips-after	off	delete each ZIP after successful extraction
--include REGEX	none	only download files whose name matches this regex (case-insensitive)
--exclude REGEX	none	skip files whose name matches this regex (case-insensitive)
--no-recursive-skip	off	disable the default recursive scan of --dest that skips files already present (matches by basename, by extracted-archive directory name, or by YYYY-MM-DD date in the filename)

Authentication

By default, HighGrabber uses the system keychain (macOS Keychain, Windows Credential Manager, or Secret Service on Linux) via the keyring package.

• highgrabber login --email you@example.com --save-password stores your password once. Subsequent logins auto-fill the form; you only touch the browser if Hightail shows a CAPTCHA.
• highgrabber logout removes the cached session; add --forget-password to also delete the keychain entry.
• A valid session lasts as long as Hightail's cookie does (usually weeks). If it expires mid-run, HighGrabber opens the browser again automatically and resumes.

How it works

HighGrabber hits three internal Hightail endpoints that the Spaces web app uses:

endpoint	purpose
GET api.spaces.hightail.com/api/v1/spaces/url/<slug>?status=SEND	resolve a receive/ slug to an internal space id
GET api.spaces.hightail.com/api/v1/files/<space_id>/untagged	list files in that space
GET download.spaces.hightail.com/api/v1/download/<sp>/<fi>/<fv>/<name>	stream one file (supports Range resume)

Authentication is a sessionId cookie on .hightail.com. HighGrabber gets it by letting you log in through a real browser (Playwright-managed Chromium), then snapshots the cookie jar to disk.

Concurrency is 2 by default. Push it higher only if you have fewer, smaller files; Hightail's rate limiter triggers reliably around 3+ parallel streams of large archives, and the symptom is HTTP 200 with an empty body.

Troubleshooting

no valid cached session / session expired prompts — run highgrabber login. Normal on a fresh install or after a long break. If you've been signed out because you logged in to Hightail from somewhere else (including another HighGrabber machine), Hightail rotates your session cookie server-side, so running login again is the fix.

The browser window closes immediately / login times out — the window waits 5 minutes for a successful login. If Hightail shows a CAPTCHA, solve it and click Sign In; the window closes as soon as the session cookie lands. If nothing appears at all, run highgrabber doctor to check that Chromium is installed.

Executable doesn't exist / Playwright browser errors — run highgrabber doctor. It installs Chromium into the right environment. The older python -m playwright install chromium pattern only works if you installed HighGrabber into your system Python, not via pipx/uv.

Downloads fail with got=0 of N or HTTP 307 redirects — you're being rate-limited by Hightail. HighGrabber already backs off (15 s / 60 s / 3 min / 5 min), but if it gives up:

• Lower --concurrency to 1.
• Wait 10–15 minutes, then re-run the same command. Partial downloads resume via HTTP Range so nothing's lost.
• If it persists for hours, your IP may be temporarily flagged; switching network (e.g. VPN on/off) usually clears it.

Wrong keychain password / authentication loop — the keyring has a stale password. Fix:

highgrabber logout --forget-password --email you@example.com
highgrabber login  --email you@example.com --save-password

macOS: keychain access denied — the macOS Keychain prompt defaults to "Allow Always" once you approve. If you clicked Deny, follow the keychain-reset recipe above.

Disk full mid-batch — partial files stay on disk; free space, re-run the same command, and downloads resume from the last byte.

Corporate proxy — HighGrabber uses httpx, which respects standard env vars:

HTTPS_PROXY=http://proxy.corp:8080 highgrabber <urls>

Windows: Chromium blocked by antivirus — some AV products quarantine Playwright's bundled Chrome Headless Shell. Allow %LOCALAPPDATA%\ms-playwright\ in your AV, then run highgrabber doctor again.

A space returns 404 / "unavailable" — Hightail expired or removed that upload. It's per-space; other links in the same run continue.

I want to see what would happen without actually downloading — highgrabber --dry-run <urls> prints every file and its size, then exits.

Ethics & Terms

HighGrabber only fetches files you already have legitimate access to via receive/<slug> links you were sent. It authenticates as you with your credentials and respects Hightail's per-session cookie model. You are responsible for your usage under Hightail's Terms of Service.

License

MIT. See LICENSE.