The baseline is the engineering reference published on 19 September 2026. Its VALIDATION.md records self-published engineering checks. The record is evidence of the stated test scope, not an independent assessment of the network. [E2][E3]
Swipe or scroll to view all columns →
| Recorded check | Result | Scope |
|---|---|---|
| Python unit and CLI checks | 9 tests passed | Validation, boundaries, Unicode, ordering and tampering |
| Python and JavaScript comparison | 80 cases passed; seed 6 | Agreement for the generated sample set |
| Extracted release quickstart | run, verify and compare passed | Local release workflow |
| Browser interactions | Execution, modified input, replay pass and tamper failure exercised | Cloud Chrome interaction |
| Layout inspection | Desktop and responsive views checked | No physical-mobile or Safari certification |
Artifact identity#
reference_runtime.py
3188b54c93976182704eef2fa4b96a4496f9aa81d965b2fd142ad550c62ea55c
reference-runtime.mjs
d116023bcf252025887cf693ad1a7d5bba92c3d4c26746b333040823d0079c24Source commit: 39d1291bc181aba31ce348b3e9f2389657618c35. The release record identifies Python 3.12.14 and Node.js 24.19.0. Checksums detect changes relative to the manifest; the manifest itself needs an authenticated distribution source.
Documentation change history#
Swipe or scroll to view all columns →
| Version | Change |
|---|---|
| Reference 0.2 | Executable local retrieval, independent replay, structural schema and release checks |
| Chinese documentation 1.0 | Architecture and network protocol review draft |
| English documentation 1.1 | Developer-oriented structure; executable English example; independent execution, acceptance and payment states; expanded recovery guidance |
Documentation 1.1 does not change the v0.2 runtime or establish network deployment. The earlier v0.1 authored receipt fixture is not an execution event generated by the v0.2 workload. Network capacity, model inference, payment and production load testing remain outside the recorded reference evidence.