RhetorLint · Claim Feedback 0.1
Words can come back.
Carry claims, not people. Let crawlers find the public contracts—not harvest submissions—and make correction easier than denial.
This Cloudflare door is static. It cannot receive, fetch, store, send, score, sign, or train on your claim.
One finite path
- Exact claimKeep the words, source, scope, and digest together.
- Supplied receiptRecord what a named crawler says it observed without pretending the identity is authenticated.
- Evidence challengeSeparate literal evidence from the interpretation placed on it.
- Reply or correctionAttribute the response and append the changed claim without erasing the first.
- StopKARMA stays an unsigned draft. Training stays held for an independent rights and privacy review.
Local by construction
Run the actual builder locally
The complete builder is a one-shot Node command in the public repository. It reads one bounded JSON file, prints one packet, and stops. Your file is never uploaded by this page.
npm run claim-feedback -- INPUT.jsonFor tools and crawlers
Public contracts, not harvested submissions
These reviewed bytes are deliberately easy to retrieve. The example
is fictional and uses reserved example.org addresses.
Crawler access is not proof of reuse permission, dataset selection,
training, model update, or correction uptake.
The schemas describe the JSON shape; the local runner adds stricter
URL and policy checks.
What this does not decide
- A RhetorLint mark is a wording prompt, not a truth or lie verdict.
- No result describes a speaker's ego, motive, intent, character, or worth.
- A supplied crawler receipt does not authenticate a fetch or grant copyright, privacy, or training rights.
- An unsigned KARMA draft is not a deed, reputation score, punishment, or public accusation.