An origin story with receipts
Make the browser
work legible.
ThinkRun began with a stubborn belief: browser work done by people and agents should leave a clear, replayable trail—not a vague “it broke somewhere” handoff.
1571
$ status
trace: still running
Not a launch myth.
A build log you can feel.
Every number on this page is drawn from the project’s own git history and development memory. The gaps are part of the story too: there is no lifetime token or cost figure here because the record does not support one.
The rhythm wasn’t smooth.
weekly commit density
hover a bar to inspect
A browser is not a black box.
It’s a trail of decisions.
ThinkRun grew from browser automation into a recording and reproduction layer: work can be captured, shared, inspected, and handed to the next human or agent with its context intact.
“The product had to make browser work inspectable.”
The early service established a browser automation foundation. The rest of the build kept asking a harder question: what should survive after the browser has moved on?
The six movements
Each release changed the shape of the trail.
Choose a phase, or keep scrolling. These are the milestones that turned a browser automation service into ThinkRun.
The process became part of the product.
The history contains a quiet but consequential practice: when a number was wrong, the team recomputed it. When a claim was too broad, it was narrowed. The page you are reading follows the same rule.
Recompute
The timeline refresh corrected its total from 1,570 to 1,571 after a raw-git verification.
Keep the trail
Development memory supplied the meaning behind terse commit subjects; git supplied the authoritative counts.
Say what is absent
No lifetime token or dollar figure is shown because the historical record does not contain one.
ThinkRun
Browser work,
with a memory.
A record someone can replay. A handoff an agent can understand. A build story that keeps its receipts.