auto-identity-remove: a scheduled runner that opts you out of data broker sites and keeps re-checking them
Automated data broker opt-out runner — removes your personal info from 30+ people-search sites on a monthly schedule
At a glance
- What is it?
- auto-identity-remove is an MIT-licensed Node and Playwright tool that searches people-search sites for your listing, submits opt-out forms on a monthly schedule, optionally solves CAPTCHAs through CapSolver, tracks state so nothing is resubmitted, and follows up with exposure scoring, CCPA and GDPR requests and breach checks.
- Who is it for?
- Take auto-identity-remove if data broker listings are a recurring problem you would rather automate than revisit, because the state tracking, the 90-day re-check window, the preview mode and the registry-fed broker list make it a maintained process rather than a one-off script, and the right-to-know and complaint commands give the legal layer most tools skip.
- Can I use it commercially?
- Yes. MIT is a permissive licence: you can use, modify and sell software built on it, as long as you keep its copyright and licence notices.
- Is it still maintained?
- Yes. The repository last received commits 37 days ago.
- What is it written in?
- Mainly JavaScript, according to GitHub's language statistics.
Answers come from the project's GitHub data, last synced on September 17, 2026, and from our analysis. They are not legal advice.
DEEP OPEN-SOURCE ANALYSIS
The monthly loop
The core job runs on a schedule and walks a list of data broker and people-search sites. For each broker it searches for your name and state, finds your specific listing where the site needs a profile URL, fills and submits the opt-out form automatically, and records the result so completed opt-outs are not resubmitted every run, with a 90-day re-check window before a broker is visited again. Forms protected by CAPTCHA are handled through CapSolver, a paid AI solving service the README prices near one tenth of a cent per solve, and without a key those sites fall back to your manual list. When the pass finishes it sends an iMessage summary, and any sites that demand a human, such as email-confirmation clicks, open in your browser. Scheduling adapts to the platform, registering a monthly job on the first at 9am through launchd, systemd, crontab or schtasks depending on what it detects.
Installing and the first run safely
The tool needs Node 18 or later and Playwright browsers, installed with one command.
npx playwright install chromiumThe quick start clones the repository, runs an install script that checks Node and installs dependencies plus Chromium, then walks an interactive setup that creates config.json and schedules the monthly job.
git clone https://github.com/stephenlthorn/auto-identity-remove.git
cd auto-identity-remove
bash install.sh
node bin/aidr.js setupBefore anything is submitted there is a preview mode, and the README positions it as a dry run that fills forms but submits nothing.
node bin/aidr.js previewOnly after checking the preview do you run the real pass.
node bin/aidr.js runSetup collects name, city, state, ZIP, email, phone, past-name aliases, a CapSolver key if you have one, accounts for sites that require login, and an iMessage number, and both config.json and state.json are gitignored so personal details stay out of any repository.
Beyond opt-outs: the aidr command set
The aidr wrapper exposes the wider loop. aidr verify re-searches brokers and reports whether you still appear, aidr score prints a 0 to 100 exposure score with a month-over-month trend built from verified removals, search-engine visibility and breach data, and aidr report generates the monthly PDF. Search surveillance comes through aidr serp and aidr serp-watch, the latter alerting only when your name appears on a new domain. The legal layer is the distinctive part: aidr know sends CCPA and GDPR right-to-know requests, and aidr complaints auto-generates regulator complaints when a broker misses the legal deadline. aidr breach checks Have I Been Pwned and recommends a credit freeze on high-severity breaches, aidr freeze walks the credit and identity-freeze checklist, aidr update-brokers refreshes the list from the official California SB 362 and Vermont broker registries, aidr pending lists sites awaiting an email confirmation click, and aidr dashboard starts a local web console with a first-run wizard, a live run view and a status board.
Privacy engineering in the tool itself
A tool whose config file holds your home address owes its users some care, and the README documents it. Submissions can go out from a masked relay email through SimpleLogin rather than your real address. config.json can be encrypted at rest with AES-256-GCM, opted in through an AIDR_PASSPHRASE environment variable. Success and confirmation text is detected in six languages, which matters because broker confirmations are not all English. The allowlist feature lets you keep brokers you deliberately want to stay listed on, and the broker list itself is sourced from public government registries rather than a hand-maintained guess. The Docker path runs everything in the official Playwright image as a non-root user with init enabled so Chromium children are reaped properly, a 512mb shared memory setting to keep renderers alive, a timezone variable because container clocks are UTC, and a low-memory mode for hosts around 2GB.
Operational limits
Three constraints shape expectations. CAPTCHA solving is a paid dependency: CapSolver is optional and without it, CAPTCHA-protected forms go to the manual list opened in your browser, or are skipped entirely with a flag, so full automation costs a small recurring amount. Coverage claims differ across the project, with the README headline saying 500 or more sites and the package description saying 30 or more, so treat the configured broker list from aidr list as the real number for your run. And the whole model depends on broker sites staying scrapable: these are third-party pages that change without notice, which is why preview and verify commands exist, and why the 90-day re-check window rather than a trust-once model is the right design. The native desktop wrapper is explicitly a planned follow-up and not included, so this remains a CLI and dashboard tool.
Compared with the alternatives
The common alternatives are manual opt-outs, which are free and precise but decay as brokers re-list people, and paid deletion services such as the commercial subscription products, which remove the labor but hold your personal data with another company. auto-identity-remove sits between them: the work runs on your machine under your keys, the personal information lives in a local, encryptable config, and the legal-tool commands go beyond what most commercial services do for the price of a CapSolver balance. The trade is that you own the maintenance, watching for broker site changes and keeping Playwright current, where a service absorbs that. For someone comfortable with a terminal who wants the process auditable, the tool is the better trade, and the readable Playwright core means every site interaction can be inspected rather than trusted.
Licence and project shape
The project is MIT licensed, versioned at 1.0.0 in package.json, and built on Playwright 1.60.0 pinned exactly, with the Dockerfile documenting that the base image tag must match the pinned Playwright version because a mismatched Chromium fails at launch, enforced by a build contract test. Dependencies are lean: axios for HTTP and nodemailer as an optional dependency, with Playwright doing the browser work. A cross-model review script in the repository suggests the author runs the code past other models, and the docs directory covers privacy, data flow, Synology NAS deployment and the low-memory mode. The command surface is complete enough that the dashboard and doctor subcommands round out a tool that expects to be operated monthly for years, which is the right frame: this is infrastructure for an ongoing privacy practice, not a one-shot cleanup script.
Editorial conclusion
Take auto-identity-remove if data broker listings are a recurring problem you would rather automate than revisit, because the state tracking, the 90-day re-check window, the preview mode and the registry-fed broker list make it a maintained process rather than a one-off script, and the right-to-know and complaint commands give the legal layer most tools skip. Budget for two things: CAPTCHA solving needs a paid CapSolver key or those forms stay manual, and the tool acts on your behalf on third-party sites, so run preview first and read the broker allowlist logic before the first real pass. The privacy design, with config encrypted at rest and a masked relay email, and the MIT licence with a readable Playwright core, make the trust question answerable, which is what a tool with your personal details in its config owes you.
Frequently asked questions
Do I need a CapSolver account to use auto-identity-remove?
No. CapSolver is optional, and without a key the CAPTCHA-protected opt-out forms are flagged as manual and opened in your browser, or skipped entirely with the no-capsolver flag.
Will the tool resubmit opt-outs for brokers I already left?
No. State tracking records completed opt-outs so they are not resubmitted each run, and brokers are re-checked on a 90-day window before being visited again.
Where does my personal information live when using this tool?
In config.json and state.json on your machine, both gitignored, with optional AES-256-GCM encryption of config.json via an AIDR_PASSPHRASE and optional SimpleLogin masked email for submissions.
Community notes