VFS Federation Studio — Engineering Handoff after v0.0.7 P1

Continue development from VFS Federation Studio v0.0.7 P1 — Federation-Scoped Monk + i18n Foundation. Do not restart, redesign, or replace the existing preservation-first architecture. Preserve all passing functionality and advance it incrementally with executable evidence.

CURRENT VERIFIED FOUNDATION

v0.0.7 preserves the complete v0.0.6 foundation: deterministic Federation Image v2 images, exact indexes, bounded embedded VFS views, recursive federation, preservation-first writable generations, generated pure-Perl runtime products, process-isolation architecture, authenticated IPC foundations, custom frameless Qt/UI contract, and executable Monk policy.

Monk profile v1 remains supported with its prior canonical serialization and behavior.

Monk profile v2 is now implemented as a backward-compatible stronger policy contract. Rules can be scoped to FEDERATION, ROOT, a specific VFS namespace, a specific mount selector, or an embedded/bundled/external DELIVERY class. VFS and mount selectors can explicitly include descendants. Scope eligibility is derived from normal federation ownership context; exact lookup remains outside Monk and remains the normal hot path.

Profile-v2 rewrites and routes fail closed when they cross an ownership/mount boundary unless the rule explicitly grants cross_mounts. Missing required ownership context produces typed VFS_E_MONK_CONTEXT; forbidden boundary crossing produces VFS_E_MONK_BOUNDARY. failure_behavior=SKIP can be used where a rule is intentionally inapplicable when required context is unavailable. Rule application, trace and path-depth budgets are enforced.

Python and generated pure-Perl runtimes now evaluate profile-v2 ownership scope and strict boundary rules with parity evidence. Writable generation rebuilding preserves the Monk profile and fingerprint.

The headless Rule Lab contract now exists. MonkRuleLab executes deterministic fixtures against the same runtime semantics and returns typed, stable results suitable for a future graphical Rule Lab. This layer must remain headless; the GUI should present worker/runtime results rather than becoming a second Monk evaluator.

Internationalization now has an architectural foundation. Stable MessageId values and a headless LocaleController keep presentation locale separate from canonical VFS semantics. The Qt adapter uses QTranslator/QCoreApplication.translate and emits locale changes for bound UI surfaces. Core shell title/open/save/build/validate strings are wired to the locale manager. Display-locale switching has executable evidence that it does not alter canonical .vfsbin bytes, including images containing permitted Unicode virtual paths.

No translation catalog for the requested four Asian languages/scripts is claimed yet. The exact four languages/scripts still require confirmation before a supported-language matrix is stated. API method names, semantic IDs, error codes, format constants and serialized policy identifiers remain language-neutral.

Generated Perl package README/API documentation now records the actual embedded Monk profile ID, schema version, fingerprint, ordered rules, scopes and v2 boundary behavior. Generated packages remain independently operational runtime products rather than demonstrations.

EXECUTABLE EVIDENCE FOR THIS CUT

Repository Python suite: 101 passed, 2 skipped. The two skips are PySide6 runtime/platform tests because PySide6 is unavailable in the assembly environment; they are not GUI passes.

Generated-runtime Perl suites: 73 assertions PASS across five test files.

P1 Monk scope/i18n proof: PASS, including nested ownership context, VFS-scoped rule selection, Rule Lab execution, explicit cross-mount write, denied strict cross-mount write, policy preservation, locale-independent deterministic Unicode image bytes, and Perl parity when Perl is available.

Preserved P0 vertical, writable-Perl and Monk proofs remain PASS.

See docs/architecture/P1_EVIDENCE.md for commands and recorded fingerprints.

PRESERVATION REQUIREMENTS

Never weaken the preservation-first model to make a feature easier. Authoritative sources remain read-only. Committed .vfsbin images are immutable generations. Mutations produce candidates, candidates are independently verified, and only the controlled commit path may promote them. Worker failure or cancellation cannot corrupt the last known-good destination. Embedded children remain bounded complete images.

Keep exact lookup out of regex policy scans. Monk is an eligibility/routing/policy layer around the normal ownership resolver, not a replacement index.

Keep Python reference behavior and generated Perl behavior in parity. A class/file existing is not proof of completion; every material claim needs executable evidence and regression coverage. Clean-extraction package tests remain mandatory. Distinguish unavailable PySide6/platform qualification from actual passes.

Keep substantial open/build/write/rebuild/search/hash/verify/export work out of the GUI event loop. The GUI presents; supervisor/workers perform substantial work. Preserve the custom frameless dialogs and application-owned SVG iconography; do not regress to native/default Qt dialogs or icons.

IMMEDIATE NEXT DEVELOPMENT

The highest-value next increment is P2: connect the proven headless Monk contracts to the professional Studio UI without duplicating policy semantics.

1. Build the graphical Rule Lab as a frameless Studio-owned engineering surface around MonkRuleLab fixtures/reports.
2. Integrate structured Monk traces into the Path Resolution Lens so exact resolution, scope eligibility, rewrites/routes, owner transitions and boundary decisions can be inspected deterministically.
3. Route nontrivial Rule Lab/trace/search evaluation through the supervisor/worker path with truthful progress/cancellation where appropriate; do not execute expensive policy/search work in the GUI event loop.
4. Continue the message-ID/i18n conversion for every new P2 surface from its first implementation. Existing untranslated legacy surfaces can be migrated incrementally, but do not introduce new hard-coded user-facing strings.
5. Add named search-profile and validation-policy contracts only through the existing Monk profile-v2 model, with Python/Perl fixtures before UI claims.
6. Preserve profile-v1 compatibility, v2 strict boundary behavior, deterministic image identity, writable Perl generation rebuilding and clean-package tests throughout P2.

The immediate UI work should consume the existing Rule Lab/runtime APIs. Do not create a UI-only regex engine or a separate trace interpretation model.

END PRODUCT CONTRACT

VFS Federation Studio is an engineering environment for creating a distinct, writable, programmable, federated virtual filesystem; exporting it as an independently operational Perl package; and proving exactly how that filesystem resolves, routes, mutates, verifies and applies policy.
