masoudmarandi's picture
meaning-continuity-lab: open reference Gradio Space
e9b473a verified
|
Raw
History Blame Contribute Delete
1.96 kB

A newer version of the Gradio SDK is available: 6.22.0

Upgrade
metadata
title: KYE Meaning Continuity Lab
emoji: 🧠
colorFrom: purple
colorTo: green
sdk: gradio
sdk_version: 5.50.0
app_file: app.py
pinned: true
license: apache-2.0
short_description: Check whether meaning survives AI agent delegation

KYE Meaning Continuity Lab™

Most systems prove what happened. KYE Protocol™ also asks whether the meaning survived.

This Space is an interactive playground for the KYE Meaning Continuity Profile™. Pick a scenario (shopping agent, finance agent, legal agent, healthcare assistant, cybersecurity agent), then change the original intent, the interpretation, the memory used, the context, the incentive context, the timing, and the handoff count.

The Space outputs:

  • a meaning-continuity score (0.0–1.0)
  • the per-dimension checked_dimensions block
  • the drift reason-codes that fired
  • the recommended admissibility / authority decision
  • a downloadable signed-style evidence pack JSON
  • the handoff trace showing where meaning changed

The scoring is an open reference implementation of the meaning continuity contract. The production runtime — KYE Meaning Continuity Engine™ and KYE Meaning Drift Monitor™ — runs the same contract at banking-grade SLAs with Ed25519-signed evidence and is part of the paid KYE Runtime layer.

Try it

  1. Choose a domain.
  2. Edit the original intent and interpretation.
  3. Toggle drift signals (memory_conflict, incentive_conflict, etc.).
  4. Watch the meaning-continuity score, reason-codes, and recommended decision update live.
  5. Download the JSON outputs to inspect or replay.

License

Apache-2.0. The contracts (schemas, vocabulary, examples) are the canonical open KYE Protocol™ contracts. The hosted engine that generates production evidence packs is commercial.

github.com/KYE-Protocol · kyeprotocol.com