geolook
Open-source end-to-end GEO implementation: status analysis, diagnosis, strategy, tickets, execution, verification
What is it?
What it is
An open-source, self-hosted platform for end-to-end Generative Engine Optimization (GEO) that helps brands get mentioned and cited by AI engines.
Why it exists
To solve the lack of visibility in AI answers, provide diagnosable root causes, turn recommendations into verifiable tickets, measure impact, and deliver client-ready GEO packages without relying on SaaS monitoring tools.
Who should use it
SEO specialistsDigital marketing agenciesBrand managersProduct teamsDevelopers comfortable with self-hosted Python apps
Who should avoid it
Users seeking a fully managed SaaS solutionTeams requiring multi-user collaboration and role-based accessThose unwilling to run command-line setup or maintain a local serverUsers needing real-time AI monitoring without manual sampling
How it works
A quick walkthrough in plain English
How geolook works
Step 1 of 3
You interact with it
Open geolook, send a request, or connect it to your stack.
Features
Advantages
- Free and open source; no subscription fees
- Complete data ownership: all data resides locally on your machine
- Full implementation loop: monitor → diagnose → tickets → assets → auto-verify → deliver
- Programmatic verification via re-crawl and next sampling round; regressions automatically reopen tickets
- Transparent, reproducible metrics anchored in public empirical data (602 prompts, 21,143 citations, etc.)
- First-class support for Chinese engines (GLM, Doubao, DeepSeek, Kimi, MiniMax, Nano, Baidu AI) and channels calibrated on citation-corpus data
- Low dependency footprint and ability to run with zero API keys using manual sampling
- Generates client-ready deliverables (diagnosis report, strategy, execution plan, ticket CSV, acceptance sheet) for agencies and consultants
- Scheduled re-runs and multi-brand switching for ongoing GEO management
- Simple security model: binds to localhost only; no built-in auth reduces attack surface
- MIT license permits unrestricted use, modification, and distribution
Disadvantages
- Designed for single-machine use; no built-in multi-user accounts or team collaboration features
- Sampling frequency and volume are limited by your own API budget (or manual effort)
- ‘Suspected negative’ flags are indicative leads requiring human review, not definitive verdicts
- Publishing actions require manual confirmation; no automated posting to live systems
- Server binds to 127.0.0.1 only; remote access necessitates SSH tunnel or reverse proxy with auth
- No integrated authentication layer; securing access relies on external measures (SSH, reverse proxy, file permissions)
- Data stored as plain files; backup relies on user-initiated git or file copies
Installation
native
git clone https://github.com/aigclink/geolook.git cd geolook pip3 install requests beautifulsoup4 lxml python3 scripts/geo.py ui
FAQ
How do I deploy GeoLook on my machine?
Clone the repo, install the three required packages (requests, beautifulsoup4, lxml), then run `python3 scripts/geo.py ui` to start the dashboard on http://127.0.0.1:8765. For remote access, use SSH port forwarding or a reverse proxy with auth.
Do I need API keys for the AI engines to use GeoLook?
No. GeoLook works with zero keys—automated sampling is skipped and you can use the manual sampling loop. Adding a single CN-capable key (e.g., DeepSeek/GLM) enables auto‑derivation of the question bank, brand facts, and AI first drafts.
How does GeoLook verify that my GEO improvements actually worked?
Verification is programmatic: after you complete a ticket, GeoLook re‑crawls your site and, in the next sampling round, checks for before/after changes. Tickets automatically reopen on regression, and verification relies on deterministic signals (site re‑crawl) rather than sampling alone.
Can GeoLook be used for the Chinese market?
Yes. GeoLook includes a first‑class Chinese engine matrix (GLM, Doubao, DeepSeek, Kimi, MiniMax, Nano, Baidu AI) and CN‑specific channels calibrated on citation‑corpus data. Metrics are measured separately for CN and global markets.
Where is my project data stored and who owns it?
All data lives locally under the `work/` directory as plain JSON/Markdown files, gitignored by design. You own everything; `git init` in `work/` serves as your backup, and there is no vendor cloud or accounts.
How does GeoLook differ from typical GEO monitoring SaaS tools?
GeoLook is an implementation platform, not just a monitor. It goes from monitoring → diagnosis → ticket generation → asset creation → auto‑verification → client‑ready deliverables, with reproducible metrics, manual publishing, and full data ownership, whereas most SaaS tools only offer dashboard screenshots and monthly subscriptions.
Featured in Videos
YouTube tutorials and walkthroughs for geolook
Alternatives
Similar projects ranked by category, topics, and text overlap.
