Journal
Dated records rather than claims: what changed, what real use taught us, and where the project's own positioning was rewritten on evidence.
Everything in this section carries a date, and every date can be checked against the repository history. That is the point of the section: a claim you cannot contradict is not worth publishing.
These are internal working records, served as they were written — the project’s own vocabulary, phase and file names included, not rewritten to read better from here. A log tidied up after the fact stops being evidence.
What each record is for
The development log records decisions, not commits — why something changed that way, and what the alternative cost. Written during the work, which is what makes it occasionally awkward: the reversals are in it too, including the evening the project’s own positioning claim was rewritten on real evidence rather than intuition.
How models behave records what language models actually do with this tool, including what they do wrong.
Web accessibility records what public sources actually answer to an agent’s discovery step — counted, not assumed. The finding →
Why the failures are here
A project that publishes only its successes tells you nothing about the day it will not work for you. This log records a real discovery gap that changed the project’s own claim about itself, a real containment leak found and fixed the same day, and the measurements that caught each of them.