Supervertaler-Workbench
CLAUDE.md62 starsChanged 5 months agoMITView on GitHub
Concise project reference for AI-assisted development. Check CHANGELOG.md for latest version and recent work. Supervertaler is a desktop CAT (Computer-Assisted Translation) tool built with PyQt6. It combines AI translation (multiple LLM providers), translation memory, terminology management, and quality assurance in a single application. Primary config: settings/settings.json with top-level sections: api_keys, general, ui, features. Satellite files under settings/: themes.json, shortcuts.json, recent_projects.json, find_replace_history.json, superlookup_history.json, voice_commands.json, model_version_cache.json. Legacy settings files are auto-migrated at startup and renamed to .migrated. 1. Supervertaler.py is very large…
1 other project uses a version of this.
Adopted by
Most recently taken up by
- michaelbeijer/Supervertalerword for word, first seen Oct 9, 2026
Other versions1 project uses a version of this.
Word for word · 1
- michaelbeijer/Supervertaler★ 62100% same
Edited · 0
None found.
Wrote their own version · 0
None found.
How matches are found. Word for word: the exact same text. Edited: shares passages with this file (“99% same” means 99% of its five-word phrases also appear here; 10% or more counts). Their own version: similar meaning, or, weakest, the same name. We can’t tell which project had it first. Projects are counted among the 200 most-starred candidates, and mentions are links to the repository we found, up to 50 per site, so both counts are a floor.
In this repository2 other files it gives its agents.
- AGENTS.mdAGENTS.md
- docs/llms.txtllms.txt
The file
We split this CLAUDE.md into the rules it gives an agent
1 rule. About 0 are common, 0 are distinctive, and 1 is very specific to this project.
≈0
Common · estimate
Many files like this have a rule of this kind.
≈0
Distinctive · estimate
This author's own choices. Worth a look.
1
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Very specific: line 66
Run python tools/shortcut_overlap.py before changing any binding: it prints the current intersection and exits non-zero if a global one …
≈ “Common” is an estimate: at least 1 in 5 of the 26,980 CLAUDE.md files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.
Files
Supervertaler – Claude Code Reference
Concise project reference for AI-assisted development. Check
CHANGELOG.mdfor latest version and recent work.
Project Overview
Supervertaler is a desktop CAT (Computer-Assisted Translation) tool built with PyQt6. It combines AI translation (multiple LLM providers), translation memory, terminology management, and quality assurance in a single application.
Key Paths
| What | Path |
| Main app (monolithic) | Supervertaler.py |
| Modules | modules/ |
| Version (source of truth) | pyproject.toml |
| Changelog | CHANGELOG.md |
| Website (landing) | docs/index.html |
| Website (Workbench page) | docs/workbench/index.html |
| Website (Trados page) | docs/trados/index.html |
| Website CSS | docs/styles.css |
| Help – Desktop (GitBook) | docs/help/supervertaler/ |
| Help – Trados (GitBook) | Moved to Supervertaler-for-Trados repo docs/ |
| Help – Desktop (old, git-synced) | docs/superdocs/ |
| Tests | tests/ |
| Settings | settings/settings.json |
| Dependencies | requirements.txt |
Important modules
modules/termlens_widget.py– TermLens inline terminology display (TermLensWidget)modules/llm_clients.py– LLM provider abstractionmodules/termbase_manager.py– glossary/termbase CRUDmodules/database_manager.py– SQLite database layermodules/translation_results_panel.py– match panel UImodules/shortcut_manager.py– keyboard shortcut systemmodules/simple_segmenter.py– sentence segmentation (SimpleSegmenter,MarkdownSegmenter)modules/platform_helpers.py– cross-platform utilities (Windows/macOS/Linux)
Settings Architecture
Primary config: settings/settings.json with top-level sections: api_keys, general, ui, features.
Satellite files under settings/: themes.json, shortcuts.json, recent_projects.json, find_replace_history.json, superlookup_history.json, voice_commands.json, model_version_cache.json.
Legacy settings files are auto-migrated at startup and renamed to .migrated.
Pitfalls
Supervertaler.pyis very large – always read/edit by line range, never full-file.- Qt table access: use
cellWidget()for editors,item()for plain items. - Block signals during programmatic text updates to avoid cascades.
- Style issues can be timing-related (hidden widgets, deferred visibility).
- XML namespace formats (SDLXLIFF): always use namespace dicts.
- Never add
"global": Trueto a shortcut whose chord Supervertaler for Trados also uses. A global hotkey is registered at OS level and fires whichever application is in front, so it reaches Trados as well – and many users run both products side by side. The two apps share 17 chords quite deliberately (Alt+↓, Ctrl+Alt+T, Alt+1…9 and so on do the same job in each), which is safe only while the Workbench side stays in-app and focus decides the winner. Promoting one of those to global turns a parallel into a collision. Two of these shipped in one day: Ctrl+Alt+A (Always-On vs Add term with abbreviation) and Ctrl+Alt+V (voice PTT vs voice toggle). The second was the nastier kind – Workbench's was a hold and Trados's a toggle, so releasing the key stopped only Workbench's and left the Trados listener latched on with nothing visible having started it. Runpython tools/shortcut_overlap.pybefore changing any binding: it prints the current intersection and exits non-zero if a global one collides.
Testing
pytest tests/Manual smoke test: import DOCX → translate → export, save/load .svproj, TM + termbase matching, AI translation, SDLPPX round-trip.
Does it work?
Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.
No reviews yet
People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.
- It worked
- Partly
- It didn't
Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.
Your trail
Files you open will collect here, so you can find them again.
Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.
Now do this for your own.
Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.

