ESP32-S3 Bench Guide
ESP32-S3 N16R8 bench guide: live pinout, lessons, simulators and the bench log.
Trace shows learning designers how a course is built: its structure and pace, where objectives are taught and checked, what learners are asked to do, and which sentences and words are hard to read.
EducationTypeScript · JavaScriptGitHub @PraxityOnlineOwner unconfirmed
Have you tried it? One tap is enough. No sign-in needed.
See how a course is built: structure and pace, where objectives are taught and checked, what learners are asked to do, and which sentences are hard to read.
Loading the full README. The saved excerpt is shown below.
Trace is a course review tool for learning designers. It reads a course and
writes one report that shows how the course is built: what each lesson is made
of, where objectives are stated, taught and checked, where the instruction for
each knowledge check sits, what learners are asked to do, where concepts appear
and return, and where the reading gets hard. Use it when you inherit a course,
before a design review, or before you rewrite a lesson. Trace shows the course;
you decide what to change.
Status: experimental. Views and the report format may change between versions.Read the full README ↗Trace is a course review tool for learning designers. It reads a course and writes one report that shows how the course is built: what each lesson is made of, where objectives are stated, taught and checked, where the instruction for each knowledge check sits, what learners are asked to do, where concepts appear and return, and where the reading gets hard. Use it when you inherit a course, before a design review, or before you rewrite a lesson. Trace shows the course; you decide what to change.
Status: experimental. Views and the report format may change between versions.
Trace needs Node 24.18 or newer and pnpm.
git clone https://github.com/Praxity/praxity-trace.git
cd praxity-trace
pnpm install
Run it with node src/cli.ts, or link the praxity-trace command with
pnpm link --global.
Trace reads three kinds of input:
praxity inspect <course-dir> --schema 1 > course.json. Schema 0 still works.praxity-trace report course.json --out report/
Open report/report.html in a browser. report/report.json holds the same data
for an LLM or a script. Every view has a Chart mode and a Table mode, and
every mark points to its lesson file and line.
Some views need judgment, such as which knowledge check covers which objective. Trace never calls an LLM itself. It prepares a bundle with the course text, a prompt and the answer format; you run it through an LLM you already use, such as Claude or ChatGPT, which writes the answer; Trace checks the answer against this revision of the course and draws the view.
praxity-trace prepare alignment course.json --out bundle/
# run bundle/prompt.md and bundle/course.md through your LLM; save its output as bundle/answer.json
praxity-trace report course.json --out report/ --alignment bundle/answer.json
The report labels these readings as interpretations. skill/SKILL.md documents every command and bundle for agents.
Praxity Trace by Ariel Harlap. Source-available under the PolyForm Perimeter License 1.0.1: free at home and at work, including paid client work. See LICENSING.md.