translate-rfq
Skill0 starsChanged 47 days agoMITView on GitHub
Translate text in an XLSX RFQ into a new workbook while preserving images, formulas, sheets, styles and other untouched archive parts. Use for customer workbooks with embedded images; translation is supplied by the agent or user, not an API.
In this repository8 other files it gives its agents.
- freight-estimatorSkill
- invoice-builderSkill
- landed-costSkill
- multi-quoteSkill
- quote-auditSkill
- supplier-rfqSkill
- whatsapp-rfqSkill
- xlsx-editSkill
The file
We split this skill into the rules it gives an agent
3 rules. About 0 are common, 0 are distinctive, and 3 are 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.
3
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Very specific: line 10
Use the standard-library scripts/translate_xlsx.py.
- Very specific: line 18
Run relative to this skill folder, or substitute the absolute script path.
- Very specific: line 30
Check the reported changed_members, untouched_members_identical, images_preserved and unfilled.
≈ “Common” is an estimate: at least 1 in 5 of the 364,980 skill 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
Preserve the RFQ workbook
If the workbook is still in a WhatsApp conversation, use the sibling whatsapp-rfq skill and the connected MCP to identify the correct chat and download the actual document. Verify the attachment's source and content, then translate that XLSX; do not replace it with a reconstructed inquiry table. This route requires both skills and an existing WhatsApp MCP connection.
Use the standard-library scripts/translate_xlsx.py. It edits only shared-string and inline-string XML text nodes, including rich-text runs. Every other archive member is verified byte-for-byte after writing. Do not round-trip the source through a spreadsheet library merely to translate it.
python scripts/translate_xlsx.py extract source.xlsx translations.json
# Fill empty values in translations.json using the requested target language.
python scripts/translate_xlsx.py apply source.xlsx translations.json translated.xlsxRun relative to this skill folder, or substitute the absolute script path. extract --glossary glossary.json optionally fills exact matches; a small ES-to-ZH demonstration glossary is supplied in assets/es-zh-demo.json. It is a sample, not a comprehensive parts dictionary.
Translation boundaries
Preserve identifiers, VINs, part numbers, brands, models, dates and numeric values. Codes containing digits are mechanically protected; names without digits require agent review. Do not add fitment, position, assembly type or quality claims absent from the source. Rich-text runs are extracted separately, so review the combined phrase in the original cell before translating them.
Empty translations keep the original visible text. They are reported as unfilled; review them before claiming a complete translation. Translation is model/user work; the script only extracts and applies supplied strings and makes no network calls.
Verification and limits
The map is bound to the source file's SHA-256. If the source changes, extract a new map. Source and destination must differ, and outputs must not already exist. XML special characters are escaped; invalid XML characters fail before publication.
Check the reported changed_members, untouched_members_identical, images_preserved and unfilled. Then open the resulting workbook and inspect representative cells, combined rich-text phrases and images. This verifies unchanged archive-member contents, not a byte-identical ZIP container.
XLSX only; encrypted, signed and duplicate-member archives are rejected. The unpacked-size limit is 512 MiB. Text in images, comments, drawings, formula results or chart labels is not translated. Publishing uses a same-directory hard link; the destination filesystem must support hard links. No source file is modified.
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.

