Zloz naming migration — local qualification
- result: complete
- claim: The four-package Zloz family, preferred public APIs, main README, demo workspace, generated API title/navigation and release identities are migrated locally. Flow-named publication is explicitly rejected; obsolete release PR #110 was closed.
- evidence: Repository check/typecheck/tests/build, workbench build, desktop/mobile editing journeys, pinned legacy-code audit, generated six-module TypeDoc browser check and production-prefix promo artifact test passed. API screenshot and browser receipt.
- remaining uncertainty: This receipt covers local qualification. Protected-PR CI and hosted acceptance are separate delivery evidence. npm publication is neither performed nor authorized. The production Effect CLI remains a subsequent slice.
- next: Review and merge the naming migration through protected CI, then verify deployed API/docs links; implement the standalone Effect CLI separately.
Verified boundaries
@opsydyn/foldkit-zloz,-elk,-forceand-mermaidinstall as renamed tarballs in clean consumers; existing Node/Bun and TypeScript qualification remains active.- Deprecated source aliases share constructor, view and cache identity with canonical APIs. Graph version 1, Mermaid
flowIdand protocol, storage keys and DOM/theme input hooks retain their spelling and behavior. - Literal version-one architecture fixtures survive editing and Undo. The pinned
ce978a2legacy reader still reads, edits, copies and round-trips the fixture. Its original Git-tree path is deliberately retained by the audit. - Canonical docs are Foldkit Zloz API, including Core, FoldKit, ELK, Force, ForceEligibility and Mermaid. Browser checks observed all six modules with no page errors. Production-path tests passed 54 assertions.
- Publication identity tests reject all four old package names and mismatched artifacts; release configuration contains only the four Zloz graph identities. No npm publish workflow was dispatched.
- Architecture/delivery/test self-review is recorded in ADR 0015. Captured historical evidence is preserved; active harness paths move with their owners. Unrelated local files are excluded.
Recommendation: merge after protected CI passes. Effect CLI (effect/cli in
Effect 4) is a recorded production requirement, not a claim that the historical
imperative fidelity driver has already been replaced.