Complete AI Training
Sign inGet my AI kit

Your job's AI kit

Get your AI kit

Tell us who you are and what you do. We show you your kit right away and email you the link: skills, prompts, AI agents, MCP servers and courses for your job.

500+ jobs ready, and we make a kit for any other job. No payment needed to look.

Share

AI agent for translators

Term Base Building Agent

Produce a consistent, approved term base for a client project.

Term Base Building Agent: what goes in, what the agent does and what you get

What it does

For a large client project, a translator often builds a term list by copying words from the source files into a spreadsheet, then looking up each translation. It is slow, and the result varies. This agent reads the client's source files and extracts candidate terms by frequency and pattern, such as product names, technical nouns and defined terms. It checks proposed translations against the client's past work, reference materials, standards and trusted glossaries, and flags conflicting entries, such as a term already used differently in an earlier project. It lists the choices for the translator to decide and, after decisions, rechecks the whole list for new conflicts. It never puts a term into use alone. The translator approves the term base before use. Edge case: a term used in two senses is split into two entries.

How it works

Follow the arrows from top to bottom. The orange dashed arrow is the loop: when a check fails, the agent goes back and tries again.

Start and resultWhat it doesA check on its own workWaits for your OKGoes back and retries
Yes, continueApprovedYes, continueNoNo 1 STARTS WHEN Project starts or files added 2 USES A TOOL Extract candidate terms from the source files 3 DOES Rank terms by frequency and importance 4 USES A TOOL Look up translations in past work, references andstandards 5 CHECKS THE RESULT Does each term have one supported translation? If not: mark the term with its options and sources fordecision. Back to step 4. 6 DOES Flag conflicts with existing entries and multi-senseterms 7 YOU APPROVE Translator decides conflicts and approves entries 8 DOES Build the term base with definitions and contexts 9 USES A TOOL Check the full list for new conflicts and duplicates 10 CHECKS THE RESULT Is the term base free of conflicts and gaps? If not: return the new conflicts to the translator. Backto step 6. 11 RESULT Approved term base
Read the steps as a list
  1. Project starts or files added
  2. Extract candidate terms from the source files
  3. Rank terms by frequency and importance
  4. Look up translations in past work, references and standards
  5. Does each term have one supported translation?If not: mark the term with its options and sources for decision. Back to step 4.
  6. Flag conflicts with existing entries and multi-sense terms
  7. Translator decides conflicts and approves entriesThe agent waits here for your OK.
  8. Build the term base with definitions and contexts
  9. Check the full list for new conflicts and duplicates
  10. Is the term base free of conflicts and gaps?If not: return the new conflicts to the translator. Back to step 6.
  11. Approved term base

How it decides

It selects a term when it appears often or is defined in the source, and recommends a translation when past approved work or a standard supports it. Conflicts go to the translator.

  • Include terms that appear 3 times or more or are defined in the source
  • Prefer the client's past approved translation
  • Split terms with two senses into two entries
  • Never merge entries with different part of speech without a check

Make it yours

Every agent is a starting point. You choose these settings for your own situation.

  • Minimum frequency (default 3)
  • Reference glossaries
  • Fields in the term base
  • Language pair
  • Export format

What keeps you in control

It always asks you first

  • Every entry
  • Resolution of conflicts

Hard limits

  • Never use a term before approval
  • Always record the source of a translation

It stops when

  • Done: term base approved
  • Stop: no reference exists for key terms

Set it up

We guide you through the set-up, step by step

Members get the full set-up guide for this agent. No technical skills needed: you copy, paste and upload.

10 minto set it up in your AI
5 AIsChatGPT, Claude, Copilot, Gemini, Grok
  • One set of instructions to paste into your AI, with the clicks for ChatGPT, Claude, Microsoft 365 Copilot, Gemini and Grok
  • The agent then walks you through connecting your own data, one source at a time
  • A downloadable copy with the flow chart, the rules and the full guide
Get access to this agent

An example run

What happensFrom 60,000 words of manuals, the agent extracted 340 candidates and ranked 180. It found support for 150 and conflicts for 11, such as bracket, which a past project translated as Halterung and Klammer in two senses. It split the entry. The translator decided 11 conflicts and approved 172 terms. A recheck found one duplicate.

More agents for translators