Drag to turn · tap a strand
A claim is only one part of the loop. The baseline, the method, and the failure make it possible to judge what happened.
Claims are cheap. Receipts are the interface.
Every entry answers a question that could have gone the other way. A baseline, a method, a measured result, what broke, and artifacts you can open. Each one is scored against six criteria. Most don't earn all six.
- ARunnable artifact
- Code, demo, or product you can open and inspect
- BMeaningful baseline
- What happens without the system — measured, not implied
- MMeasured result
- A number that could have come out worse
- RReproducible method
- Steps or harness another engineer can rerun
- FDocumented failure
- Where it still broke, on the record
- EExternal validation
- Someone who is not me depends on it or verified it
Receipts, newest first
7 receipts on the ledger · next drafted daily at 07:10 · external validation earned by 2 of 7
A portable contract that rejects missing authority
Can another runtime validate what an agent was asked to do, what it was authorized to do, what changed, and how the result was checked — without an OrgX account?
Where it broke: The package is still a repository preview and is not published to npm.
A solo Rust tool that installs like a real product
Can one person ship a native developer tool with the full distribution surface — versioned releases, checksums, a Homebrew tap, a landing page — not just a repo?
One persistent signal: the thread survives the route handoff
Can a site's core claim — intelligence survives the handoff — be made structurally true: one visual element that literally never unmounts across navigation?
57 governed tools, 654 tests, one live MCP server
Can one MCP server carry organizational memory, bounded authority, and receipts to every major AI client — and stay up in production?
12 tasks, 7 domains, 3 execution modes — with a human baseline
Does orchestrated multi-agent execution actually beat a single agent — and a human — on real initiative-level work, measured the same way every week?
Eight motion packages, live on public npm
Is the 'token-driven Remotion monorepo' a real, installable system — or a private folder with a nice README?
Config 2021: Figma platformed the community I built
Did the community-through-making thesis — less talk, more people in the file — hold up in front of the industry it came from?
The rule here: four of six criteria before an entry ships, and the failure field is never empty. External validation is the hardest one to earn. That's why it's tracked.
Inspect the systems behind the receipts