# LexiconError v1.0.0

## Stable desktop release

LexiconError v1.0.0 is the first stable release of the offline diagnostic encyclopedia. It combines a 16,474-record SQLite/FTS5 catalog with an information-rich native desktop workspace for understanding compiler diagnostics, runtime exceptions, infrastructure failures, and GPU runtime faults.

## Highlights

- 16,474 locally searchable records across 44 ecosystems, with no account, server, telemetry, or required network connection.
- Four interactive inspector workspaces for root-cause analytics, trigger-versus-fix code, situational context and cross-language relationships, and failure-state provenance.
- Experience-likelihood visualization that explicitly distinguishes curated frequency from measured incident probability.
- Catalog-derived frequency and failure-domain graphs with clickable, ecosystem-scoped filters.
- Resolvable related-diagnostic navigation across languages, including conceptual equivalents.
- Verification state, toolchain identity, canonical IDs, version boundaries, catalog cohort counts, source quality, and dataset provenance.
- SQLite FTS5 search, pasted-error detection with sensitive-line redaction, hierarchical filters, command palette, context matrix, and coverage dashboard.
- Validated JSON and Markdown/YAML contribution imports plus portable SQLite backup and structured export utilities.
- CUDA, ROCm/HIP, Vulkan, and OpenCL coverage alongside core language, framework, infrastructure, database, and configuration diagnostics.

## Desktop experience

The v1.0.0 interface replaces the earlier thin inspector with a denser three-pane diagnostic workspace. Every visible analytic control performs a real local query or filter action. Frequency charts use catalog counts, related-diagnostic cards open actual indexed records, and the provenance view reports the verification and dataset metadata stored with the selected diagnostic.

## Release assets

- `LexiconError_1.0.0_x64-setup.exe` - recommended Windows x64 installer.
- `LexiconError_1.0.0_x64_en-US.msi` - MSI package for managed Windows deployment.
- `lexicon-error-2026.08-reviewed.db` - portable SQLite/FTS5 catalog backup.

The desktop installers bundle the reviewed catalog and remain fully functional offline after installation.

## SHA-256 checksums

~~~text
9984E765EE38616834CE6C8DD88987D498621E2E01B6BC0815E65FCEE3C5A445  LexiconError_1.0.0_x64-setup.exe
55DB308BFAB62DA6AC446F830D495763DCE0DB7991B7911E048739B4763C5C65  LexiconError_1.0.0_x64_en-US.msi
D54491F4D5D02C8034E0664B6F1198627697440D9CE57AB297738E70B764CF14  lexicon-error-2026.08-reviewed.db
~~~

## Verification performed

~~~text
npm.cmd run check
npm.cmd run desktop:build
cargo test --manifest-path src-tauri\Cargo.toml
python ingestion\test_pipeline.py --database artifacts\lexicon-error-2026.08-reviewed.db
python ingestion\test_hf_package.py --package hf\lexiconerror-diagnostics --space hf\lexiconerror-space --pages site
python -m py_compile scripts\capture_readme_media.py
~~~

The compiled app was also exercised at 1440 x 900 while regenerating all six screenshots, the eight-frame GIF, and the 45-frame H.264 demonstration video.

## Important notes

- The included native binaries target Windows x64. Build the same source on macOS or Linux to create platform-native bundles.
- Catalog volume does not imply that every generated registry record is editorially complete. Use the visible verification state and official source link when accuracy matters.
- Frequency labels are curated ordinal classifications, not live telemetry or guaranteed real-world probabilities.
- Code examples are documentation and are never executed by LexiconError.
- Local contributions are stored in the user application-data database and are not added automatically to the bundled portable catalog.

## Data and support

The public derived-metadata dataset preserves provenance and source notices. Its aggregate license is `other` because upstream terms differ; review the repository source manifest and dataset notice before republishing the complete corpus.

If LexiconError helps your work, consider supporting `@theworker02` through the repository Sponsor button once GitHub Sponsors is active.