Tools
Measured, not estimated.
All tools disclose their assumptions and work with real data.
-
Available
QuickScan
Checks for free how well search engines and AI agents can read a domain and how well it is secured. The score appears on screen right away. On request it also arrives by email, with the three most important recommendations.
Free, no account. Ten requests per minute and IP. Runs on validator.litzki-systems.com.
-
Available
SEO Forecast Calculator
Forecasts clicks from search volume, position and device, with CTR curves whose source you can see.
Runs in your browser on this site, without cookies. Open source (MIT).
-
Available
Briefing Generator
12 free briefing templates: fill them in your browser and hand them to your agency or writer as a fillable PDF.
Runs in your browser on this site, without cookies. Open source (MIT).
-
In preparation
Linkbuilding Architect
Plans and documents link building with a traffic gap calculator, templates and disavow export, local and open source.
Open source on GitHub, link to follow.
-
Available
Prompt Library
Reviewed prompts and style rules for SEO copy, briefs and audits, as Claude Skills. First skill: SEO-Content-Workflow for German-language SEO copy, seven checked stages from brief to CMS package, with no invented experience, sources or metrics. Passed 34 of 34 checks in testing, 23 without the skill.
Open source (MIT). More skills to follow.
-
In use · internal
Keyword Command Center
Keyword research with measured search volume instead of AI estimates. Analysis, import, cluster mapping and forecast through to the client report – every number with its source. Internal Litzki Systems tool, used in my client projects.
Sign-in only, for Litzki Systems projects. Numbers from DataForSEO, CSV import and Search Console; the AI never supplies numbers.
-
Available
infra-check
Checks a domain on five machine-readable signals: JSON-LD, llms.txt, robots.txt for AI crawlers, security headers and TLS. Runs on the command line, also for whole domain lists.
Open source (MIT).
-
Available
chatgpt-render-check
Simulates how ChatGPT reportedly splits a fetched page into text chunks and measures across 1,000 domains how often a plain text conversion cuts elsewhere: on 363 of 376 multi-chunk pages.
Open source (MIT). An openly documented simulation.