30 Claude Prompts for Translators
Paste in your source text, glossary, or client brief and get back a usable draft: a terminology table, a QA pass with flagged inconsistencies, or a localisation brief you can hand to a client.
In short: This page contains 30 copy-paste ready prompts, organized into 6 categories with a description and pro tip for each. The first 5 prompts are free instantly, no signup needed. Hand-curated and tested by the AI Academy team.
Glossary Building
5 promptsBuild a bilingual glossary from a source document
1/30โจ What it does
Produces a starter bilingual glossary table with confidence ratings from a source document.
You are a senior terminologist who builds bilingual glossaries for translation teams. <context> I have a source document that will be translated into another language and I need a starting glossary before translation begins. </context> <inputs> - Source language: [SOURCE LANGUAGE] - Target language: [TARGET LANGUAGE] - Domain: [DOMAIN, E.G. MEDICAL DEVICES] - Source text or excerpt: [PASTE SOURCE TEXT] - Existing termbase, if any: [PASTE EXISTING TERMS OR WRITE NONE] </inputs> <task> Extract the key terms, product names, and recurring phrases from the source text and propose a target-language equivalent for each one, flagging any term that has more than one common translation. </task> <constraints> Limit the glossary to terms that appear more than once or that carry domain-specific meaning. Do not translate generic words. Note where a literal translation would be misleading. Keep proposed equivalents short, no full sentences. </constraints> <format> Return a table with columns: Source term, Proposed target term, Domain note, Confidence (high, medium, low). Add a short list underneath of terms that need client confirmation. </format>
Pro tip: Run this before you start the actual translation so the glossary is locked before terminology drift creeps in.
Reconcile two glossaries after a merger or handoff
2/30โจ What it does
Merges two conflicting glossaries into one recommended termbase with reasons per decision.
You are a terminology manager who consolidates termbases when two translation teams merge their work. <context> I have two glossaries for the same client that were built by different translators and they now conflict on several terms. </context> <inputs> - Glossary A: [PASTE GLOSSARY A] - Glossary B: [PASTE GLOSSARY B] - Target language: [TARGET LANGUAGE] - Client preference, if known: [CLIENT PREFERENCE OR WRITE UNKNOWN] </inputs> <task> Compare both glossaries term by term, flag every conflict, and recommend which version to keep based on frequency of use, consistency with the domain, and any stated client preference. </task> <constraints> Do not silently pick a winner without stating the reason. Group conflicts by severity, meaning terms that change client-facing meaning versus terms that are just stylistic variants. Keep the reasoning to one line per term. </constraints> <format> Return a merged table with columns: Term, Glossary A version, Glossary B version, Recommendation, Reason. End with a short list of terms that need a decision from the client. </format>
Pro tip: Send the final list of undecided terms straight to the client as a short approval email instead of guessing.
Turn a style guide into a do-not-translate list
3/30โจ What it does
Extracts a do-not-translate list split into hard and soft rules from a client style guide.
You are a localisation specialist who prepares do-not-translate lists for CAT tool setup. <context> I am setting up a translation memory tool and I need a clean list of terms that must stay in the source language, such as brand names and product SKUs. </context> <inputs> - Source language: [SOURCE LANGUAGE] - Target language: [TARGET LANGUAGE] - Client style guide or brand list: [PASTE STYLE GUIDE TEXT] - Product or company name: [PRODUCT OR COMPANY NAME] </inputs> <task> Scan the style guide and pull out every term, name, or phrase that should not be translated, and explain briefly why each one stays in the source language. </task> <constraints> Separate hard rules, meaning legally protected names, from soft rules, meaning terms the client just prefers untranslated. Do not include generic vocabulary. Flag any term that is ambiguous, meaning it could be a brand name or a common word depending on context. </constraints> <format> Return two lists: Hard rules (must never translate) and Soft rules (client preference), each with the term and a one-line reason. Add a short note on ambiguous terms. </format>
Pro tip: Load the hard rules list directly into your CAT tool's non-translatable terms setting before you start the job.
Draft a termbase entry with usage examples
4/30โจ What it does
Writes a complete, ready-to-load termbase entry for a single problem term.
You are a terminologist who writes detailed termbase entries for enterprise translation memory systems. <context> I need a full termbase entry for a single technical term because it keeps causing confusion across translators on this account. </context> <inputs> - Term: [TERM] - Source language: [SOURCE LANGUAGE] - Target language: [TARGET LANGUAGE] - Domain: [DOMAIN] - Sentence where the term appears: [PASTE SENTENCE] </inputs> <task> Write a complete termbase entry for this term, including the approved target translation, a short definition, a usage example in both languages, and any forbidden alternate translations. </task> <constraints> Keep the definition to two sentences maximum. The usage example must be a natural sentence, not a fragment. List at most three forbidden alternates with a one-line reason each. Avoid jargon in the definition itself. </constraints> <format> Return as a labeled entry: Term, Approved translation, Definition, Usage example (source), Usage example (target), Forbidden alternates with reasons. </format>
Pro tip: Save these entries in a running document per client so new translators on the account inherit the reasoning, not just the term.
Spot missing terms by comparing a translated file to its glossary
5/30โจ What it does
Flags glossary inconsistencies and possible omissions in a finished translation.
You are a QA terminologist who checks translated documents against an approved glossary. <context> I have a finished translation and an approved glossary and I need to confirm the translation actually used the approved terms consistently. </context> <inputs> - Approved glossary: [PASTE GLOSSARY] - Translated text: [PASTE TRANSLATED TEXT] - Target language: [TARGET LANGUAGE] </inputs> <task> Check the translated text against the glossary and list every place where a glossary term was translated inconsistently or with an unapproved alternative. </task> <constraints> Quote the exact sentence where the mismatch occurs. Do not flag stylistic paraphrasing that does not affect a glossary term. If a term is missing entirely from the translation despite appearing in the source implied context, note that as a possible omission for human review. </constraints> <format> Return a table with columns: Glossary term, Approved translation, Found in text, Location or quote, Issue type (inconsistent, unapproved, possibly omitted). </format>
Pro tip: Run this as a final pass before delivery, it catches drift that happens naturally over a long document.
Localisation QA
5 promptsRun a linguistic QA pass on a translated UI string set
6/30โจ What it does
Checks a batch of translated UI strings for placeholder errors, length risk, and tone issues before release.
You are a localisation QA specialist who checks translated software strings before release. <context> I have a batch of translated UI strings and I need a quality check before they ship in the next app release. </context> <inputs> - Source strings with keys: [PASTE SOURCE STRINGS WITH KEYS] - Translated strings with keys: [PASTE TRANSLATED STRINGS WITH KEYS] - Target language: [TARGET LANGUAGE] - Platform: [PLATFORM, E.G. IOS, ANDROID, WEB] </inputs> <task> Check each translated string against its source for meaning accuracy, placeholder integrity such as %s or {name}, truncation risk on short UI elements, and tone consistency with a product interface. </task> <constraints> Flag any string where a placeholder or variable was dropped or altered, that is a release blocker. Note strings that are likely too long for a button or label. Do not rewrite the strings yourself, only flag issues with a suggested fix. </constraints> <format> Return a table with columns: Key, Issue type (accuracy, placeholder, length, tone), Description, Suggested fix, Severity (blocker, major, minor). </format>
Pro tip: Filter the output to blocker severity first, that is the set that must be fixed before you can approve the build.
Compare back-translation to the original for a legal clause
7/30โจ What it does
Compares a legal clause to its back-translation and flags any shift in obligations or scope.
You are a legal translation reviewer who verifies meaning through back-translation. <context> I translated a contract clause and had it back-translated by another linguist, and I need to confirm the meaning survived both passes intact. </context> <inputs> - Original clause: [PASTE ORIGINAL CLAUSE] - Back-translated clause: [PASTE BACK TRANSLATED CLAUSE] - Source language: [SOURCE LANGUAGE] - Governing law or jurisdiction, if relevant: [JURISDICTION OR WRITE NONE] </inputs> <task> Compare the meaning, obligations, and conditions in the original clause against the back-translated version and identify any shift in legal meaning, scope, or obligation. </task> <constraints> Focus only on substantive meaning, not wording style. Call out any change to who is obligated, under what condition, or with what deadline. If the clause is ambiguous in either version, say so explicitly rather than guessing intent. </constraints> <format> Return a short verdict line, meaning matches or meaning has shifted, followed by a table with columns: Element (obligation, condition, deadline, party), Original meaning, Back-translation meaning, Match (yes, no, unclear). </format>
Pro tip: Always route a flagged mismatch to a qualified legal translator, this prompt narrows the search, it does not replace legal review.
Check number, date, and currency formatting across a localised document
8/30โจ What it does
Audits number, date, and currency formatting in a translated document against target locale rules.
You are a localisation QA reviewer who specializes in locale formatting conventions. <context> I have a translated document and I need to confirm every number, date, and currency value follows the target locale's conventions, not the source locale's. </context> <inputs> - Translated text: [PASTE TRANSLATED TEXT] - Source locale: [SOURCE LOCALE, E.G. EN-US] - Target locale: [TARGET LOCALE, E.G. FR-FR] </inputs> <task> Scan the text for every number, date, time, and currency value and check whether it follows target locale formatting conventions such as decimal separators, date order, and currency symbol placement. </task> <constraints> List each value found, not just the ones that are wrong, so I can spot-check the correct ones too. Do not change the actual numeric value, only its formatting. Flag ambiguous dates where the day and month could be confused between locales. </constraints> <format> Return a table with columns: Original value, Found format, Expected target format, Status (correct, incorrect, ambiguous). </format>
Pro tip: Ambiguous dates like 03/04 are the most common silent error, always confirm those with the client rather than assuming.
Identify cultural or idiomatic red flags before publishing
9/30โจ What it does
Flags idioms, cultural references, and tone mismatches in translated marketing copy before launch.
You are a cultural localisation consultant who reviews translated marketing copy for the target market. <context> I have translated marketing copy and I want a check for anything that could read badly, confuse, or offend in the target culture before it goes live. </context> <inputs> - Translated copy: [PASTE TRANSLATED COPY] - Target country and language: [TARGET COUNTRY AND LANGUAGE] - Product or campaign context: [PRODUCT OR CAMPAIGN CONTEXT] </inputs> <task> Review the copy for idioms that do not translate well, cultural references that will not land, imagery or claims implied by the text that could be sensitive locally, and tone mismatches with local marketing norms. </task> <constraints> Be specific about why something is a risk, not just that it feels off. Distinguish between a hard problem that needs a rewrite and a minor stylistic note. Do not invent cultural rules you are not confident about, say when you are unsure and recommend a native reviewer instead. </constraints> <format> Return a table with columns: Excerpt, Concern, Risk level (high, medium, low), Recommendation. End with a one-line overall verdict on whether native review is needed. </format>
Pro tip: Treat high risk items as mandatory native reviewer sign-off, this prompt is a first filter, not the final word on culture.
Build a QA checklist tailored to one language pair
10/30โจ What it does
Builds a reusable, checkable QA checklist tuned to a specific language pair and content type.
You are a quality assurance lead who builds recurring QA checklists for translation teams. <context> I run QA for a specific language pair repeatedly and I want a standing checklist that captures the errors this pair tends to produce, so reviewers stop missing the same things. </context> <inputs> - Source language: [SOURCE LANGUAGE] - Target language: [TARGET LANGUAGE] - Content type: [CONTENT TYPE, E.G. SOFTWARE, LEGAL, MARKETING] - Recurring issues seen so far: [LIST RECURRING ISSUES OR WRITE NONE YET] </inputs> <task> Build a QA checklist specific to this language pair and content type, covering grammar patterns that commonly break in translation, formatting differences, and any recurring issues I listed. </task> <constraints> Keep each checklist item actionable, meaning a reviewer can check yes or no against it, not vague advice. Group items by category such as grammar, formatting, terminology, and tone. Limit to the most common failure points, not an exhaustive grammar reference. </constraints> <format> Return a checklist grouped by category, each item as a short checkable statement. Add a note at the end on how to extend the list as new recurring issues appear. </format>
Pro tip: Keep this checklist as a living document and add one line every time a client rejects something for a new reason.
Style Guides
5 promptsDraft a translation style guide from a client's brand voice
11/30โจ What it does
Converts a client's source-language brand voice guidelines into concrete style guidance for the target language.
You are a translation style guide author who converts brand voice documents into working style guides for translators. <context> A new client sent me their brand voice guidelines in the source language and I need a translation style guide that captures how their voice should carry into the target language. </context> <inputs> - Source brand voice guidelines: [PASTE BRAND VOICE TEXT] - Source language: [SOURCE LANGUAGE] - Target language: [TARGET LANGUAGE] - Content types covered: [CONTENT TYPES, E.G. WEBSITE, EMAILS, ADS] </inputs> <task> Translate the intent of the brand voice, not the literal wording, into concrete guidance for a translator working in the target language, covering tone, formality level, sentence length preference, and words or phrases to avoid. </task> <constraints> Do not just translate the source guidelines word for word, adapt the guidance so it makes sense as instructions for the target language. Include at least three concrete example sentences showing the tone in the target language. Keep formality guidance explicit, since this varies more across languages than in the source. </constraints> <format> Return sections: Tone summary, Formality level, Sentence style, Words and phrases to avoid, Three example sentences in the target language. </format>
Pro tip: Send the three example sentences back to the client for a quick yes or no before applying the style guide to the full project.
Reconcile client style feedback into a single rule
12/30โจ What it does
Turns scattered client feedback comments into a single clear, reusable style guide rule.
You are a translation project manager who converts scattered client feedback into style guide rules. <context> A client has sent feedback on several delivered translations and I need to turn their comments into a clear, reusable style rule for future work. </context> <inputs> - Client feedback comments: [PASTE CLIENT FEEDBACK] - Original translated sentences the feedback refers to: [PASTE ORIGINAL SENTENCES] - Target language: [TARGET LANGUAGE] </inputs> <task> Read the feedback and the sentences it refers to, identify the underlying pattern behind the client's preference, and write it as a single clear style guide rule with a correct and incorrect example. </task> <constraints> If the feedback is contradictory or too vague to turn into a rule, say so directly and list what clarifying question to ask the client. Do not invent a rule broader than what the feedback actually supports. </constraints> <format> Return: Rule statement, Correct example, Incorrect example, Confidence (high, medium, needs clarification), and if needed a clarifying question for the client. </format>
Pro tip: Batch several feedback comments into one run so you build the style guide in a handful of rules instead of one at a time.
Write formality and tone guidance for a new market
13/30โจ What it does
Produces formality and tone guidance for a specific target market and content type.
You are a localisation consultant who advises on formality conventions for new target markets. <context> A client is expanding into a new target language market and wants guidance on how formal or casual the translated content should be for that market and content type. </context> <inputs> - Target language and country: [TARGET LANGUAGE AND COUNTRY] - Content type: [CONTENT TYPE, E.G. B2B SOFTWARE, CONSUMER APP] - Source tone in original language: [SOURCE TONE DESCRIPTION] - Audience: [AUDIENCE DESCRIPTION] </inputs> <task> Recommend a formality level and tone approach for this market and content type, explain how it differs from the source tone if it does, and give guidance on pronoun choice, sentence structure, and any honorific conventions relevant to the language. </task> <constraints> Be specific to the named country, not the language in general, since formality norms can vary between countries that share a language. Flag anywhere you are giving a general recommendation rather than certainty, so a native reviewer can confirm. </constraints> <format> Return sections: Recommended formality level, Reasoning, Pronoun and honorific guidance, Two example sentences, Confidence note. </format>
Pro tip: Confirm the recommendation with a native speaker on the client team before locking it into the style guide, formality norms shift fast.
Convert an English style guide into target-language grammar rules
14/30โจ What it does
Maps an English style guide rule by rule to what actually applies in the target language.
You are a grammar and style specialist who adapts English style rules for other languages. <context> I have an English-language editorial style guide from a client and I need to know which rules actually apply once translated and which ones need a different equivalent in the target language. </context> <inputs> - English style guide excerpt: [PASTE ENGLISH STYLE GUIDE] - Target language: [TARGET LANGUAGE] - Content type: [CONTENT TYPE, E.G. LEGAL, MARKETING, SOFTWARE] </inputs> <task> Go through each rule in the English style guide and state whether it transfers directly to the target language, needs adaptation, or does not apply at all, and propose the target-language equivalent where adaptation is needed. </task> <constraints> Do not assume English punctuation or capitalization rules apply elsewhere, check each one. Where a rule does not apply, explain briefly why the target language handles it differently. Keep the target-language equivalents concrete enough for a translator to apply directly. </constraints> <format> Return a table with columns: English rule, Status (transfers, needs adaptation, does not apply), Target language equivalent or note. </format>
Pro tip: Pay closest attention to punctuation and quotation mark rules, they are the most commonly copied over by mistake.
Create a quick-reference card from a long style guide
15/30โจ What it does
Condenses a long style guide into a one-page quick-reference card focused on commonly missed rules.
You are a documentation specialist who condenses long style guides into quick-reference material for translators. <context> I have a long style guide document that translators on my team rarely read in full, and I want a one-page quick reference they will actually use during a job. </context> <inputs> - Full style guide text: [PASTE FULL STYLE GUIDE] - Target language: [TARGET LANGUAGE] - Most common mistakes translators make on this account: [LIST COMMON MISTAKES OR WRITE NONE YET] </inputs> <task> Condense the style guide into a short quick-reference card that highlights the rules most likely to be missed or forgotten, prioritizing anything tied to the common mistakes I listed. </task> <constraints> Limit the result to what fits comfortably on one page. Cut anything that is obvious or standard practice and would not need reminding. Order items by how often they are likely to matter, most important first. </constraints> <format> Return a short bulleted quick-reference card, grouped under two or three headings, each bullet one line long. </format>
Pro tip: Pin this quick reference at the top of the shared drive folder, a full style guide that nobody opens is not doing its job.
These prompts give you the what. Tutorials give you the why.
Learn when to use extended thinking, how to build Claude Projects, and workflows that compound. 300+ tutorials and growing.
Transcreation Briefs
5 promptsWrite a transcreation brief from a source ad campaign
16/30โจ What it does
Produces a transcreation brief that separates fixed campaign elements from what a translator can freely adapt.
You are a transcreation strategist who prepares briefs for adapting ad campaigns across markets. <context> A client's ad campaign needs to be adapted, not literally translated, for a new market, and I need a brief that gives the creative translator enough context to do that well. </context> <inputs> - Source campaign copy: [PASTE SOURCE CAMPAIGN COPY] - Campaign objective: [CAMPAIGN OBJECTIVE] - Target market: [TARGET MARKET] - Constraints from the client: [CLIENT CONSTRAINTS OR WRITE NONE] </inputs> <task> Write a transcreation brief that explains the emotional intent and key message of the source campaign, flags any wordplay or cultural reference that will not survive literal translation, and sets out what must stay fixed versus what can be freely reimagined for the target market. </task> <constraints> Separate must-keep elements, such as the offer or legal disclaimer, from free-to-adapt elements, such as tone or wordplay. Do not attempt the transcreation yourself in this brief, only set up the assignment. Keep the brief usable by a translator who has not seen the original campaign meeting. </constraints> <format> Return sections: Core message and emotional intent, Elements that must stay fixed, Elements open to adaptation, Cultural or wordplay risks, Success criteria for the target market version. </format>
Pro tip: Get the fixed elements list signed off by the client before the transcreator starts, that is the section most likely to cause revision loops.
Generate three transcreation directions for a tagline
17/30โจ What it does
Generates three distinct creative directions for adapting a tagline, each with its own mechanism and risk.
You are a creative transcreation specialist who proposes multiple directions for adapting taglines across languages. <context> I need to adapt a tagline for a new market and I want a few different creative directions before committing to one, since a literal translation will not carry the wordplay. </context> <inputs> - Source tagline: [SOURCE TAGLINE] - What the wordplay or hook relies on in the source: [EXPLAIN THE HOOK] - Target language: [TARGET LANGUAGE] - Brand tone: [BRAND TONE] </inputs> <task> Propose three distinct transcreation directions for the tagline in the target language, each preserving the emotional effect of the original hook through a different mechanism, such as rhythm, local idiom, or a new wordplay native to the target language. </task> <constraints> Do not translate the tagline literally as one of the three options, all three must be genuine adaptations. For each direction, explain in one sentence what mechanism it uses to recreate the effect. Keep each tagline short enough to work as an actual tagline, not a sentence. </constraints> <format> Return three options, each with: Proposed tagline, Mechanism used, One risk to flag. End with a one-line recommendation of which to test first. </format>
Pro tip: Test the top two directions with a small native-speaker panel before final sign-off, taglines are the highest-risk single line in a campaign.
Assess whether a phrase needs transcreation or straight translation
18/30โจ What it does
Triages a mixed document into sections needing creative transcreation versus standard translation.
You are a localisation triage specialist who decides which parts of a document need creative transcreation versus straightforward translation. <context> I have a mixed document with marketing copy and factual content, and I want to know which parts actually need a creative transcreation pass versus which can go through standard translation. </context> <inputs> - Full document text: [PASTE DOCUMENT TEXT] - Target language: [TARGET LANGUAGE] - Budget or time constraint, if relevant: [CONSTRAINT OR WRITE NONE] </inputs> <task> Go through the document and classify each section or sentence as needing transcreation, meaning it relies on wordplay, humor, or emotional hook, or as suitable for standard translation, meaning it is factual or instructional. </task> <constraints> Be conservative, only flag transcreation where a literal translation would genuinely lose meaning or impact, not for every stylistic sentence. Explain the reason for each transcreation flag in one line. Group the output so I can route work efficiently. </constraints> <format> Return two lists: Needs transcreation (with excerpt and reason) and Standard translation is fine (just list the sections). End with a rough percentage split. </format>
Pro tip: Use the percentage split to quote transcreation work separately and accurately instead of pricing the whole document at one rate.
Write a back-brief summary to confirm transcreation intent with the client
19/30โจ What it does
Drafts a short client-facing back-brief that confirms creative intent before transcreation work begins.
You are a client-facing localisation manager who confirms creative intent before starting transcreation work. <context> Before I start a transcreation job I want to send the client a short back-brief so they confirm I understood the intent correctly, to avoid a costly revision cycle later. </context> <inputs> - Client's original brief or notes: [PASTE CLIENT BRIEF] - Source material to be transcreated: [PASTE SOURCE MATERIAL] - Target market: [TARGET MARKET] </inputs> <task> Write a short back-brief summary that restates my understanding of the creative intent, the target audience, the tone, and the constraints, and ends with two or three specific questions to resolve before I start. </task> <constraints> Keep the summary short enough that a busy client will actually read it, no more than half a page. The questions must be specific and answerable in one line each, not open-ended. Do not include transcreation options yet, this is a confirmation step only. </constraints> <format> Return: Summary of understood intent (short paragraph), Target audience and tone, Confirmed constraints, Two to three specific questions. </format>
Pro tip: Send this before any creative work starts, it is far cheaper to fix a misunderstanding here than after three drafts.
Evaluate a completed transcreation against the original brief
20/30โจ What it does
Checks a finished transcreation against its original brief and flags whether must-keep elements and tone were delivered.
You are a quality reviewer who checks finished transcreation work against its original creative brief. <context> A transcreation has come back from a translator and I need to check it actually delivers against the brief before sending it to the client. </context> <inputs> - Original transcreation brief: [PASTE BRIEF] - Delivered transcreated copy: [PASTE DELIVERED COPY] - Target market: [TARGET MARKET] </inputs> <task> Compare the delivered copy against the brief's stated must-keep elements, tone, and success criteria, and assess whether it meets the brief, partially meets it, or misses it, with reasons. </task> <constraints> Check the must-keep elements first since those are non-negotiable, then assess the more subjective creative fit. Do not rewrite the copy, only evaluate it. If the copy deviates from the brief but the deviation seems like a legitimate creative improvement, note that as a judgment call for the client rather than a fail. </constraints> <format> Return: Must-keep elements check (pass or fail per element), Tone and creative fit assessment, Overall verdict (meets brief, partial, misses brief), Notes for the client if a deviation looks intentional and good. </format>
Pro tip: Run this before you forward anything to the client, catching a missed must-keep element internally saves an awkward revision request.
Terminology Management
5 promptsDetect inconsistent terminology across multiple translated files
21/30โจ What it does
Detects the same source term translated inconsistently across multiple delivered files.
You are a terminology auditor who checks consistency across a batch of translated files for the same client. <context> I have several translated files delivered over time for the same client and I suspect the same source term has been translated differently across them. </context> <inputs> - File 1 excerpt: [PASTE FILE 1 EXCERPT] - File 2 excerpt: [PASTE FILE 2 EXCERPT] - File 3 excerpt, if any: [PASTE FILE 3 EXCERPT OR WRITE NONE] - Target language: [TARGET LANGUAGE] </inputs> <task> Find terms that appear to refer to the same source concept but were translated differently across the files, and list them so I can decide which version to standardize on. </task> <constraints> Only flag genuine terminology drift, not natural variation in ordinary sentences. Quote the exact sentence each variant appears in so I can trace it back. Do not decide which version is correct, only surface the inconsistency. </constraints> <format> Return a table with columns: Likely source concept, Variant found, File it appears in, Quoted sentence. </format>
Pro tip: Run this every few deliveries on long-running accounts, drift creeps in gradually and is easy to miss file by file.
Propose a target-language term for a newly coined source term
22/30โจ What it does
Proposes and reasons through candidate target-language translations for a newly coined source term.
You are a terminology creator who proposes target-language terms for new concepts that do not yet have an established translation. <context> The source document uses a term that is newly coined or very niche, and there is no established translation for it in the target language yet, so I need a defensible proposal. </context> <inputs> - New term: [NEW TERM] - Definition or context of use: [DEFINITION OR CONTEXT] - Target language: [TARGET LANGUAGE] - How formal or technical the audience is: [AUDIENCE DESCRIPTION] </inputs> <task> Propose one or two candidate translations for the new term, explain the reasoning behind each, such as whether it is a direct coinage, a descriptive phrase, or a borrowed loanword, and recommend which fits the audience best. </task> <constraints> Check whether a loanword is already common in the target language for similar concepts before proposing one. Flag if the term is likely to need eventual standardization by an industry body rather than a one-off translator decision. Keep the reasoning grounded in how the target language actually forms new technical terms, not a guess. </constraints> <format> Return: Candidate 1 with reasoning, Candidate 2 with reasoning if applicable, Recommendation, Note on whether this needs wider industry confirmation. </format>
Pro tip: Flag your proposal clearly as a working translation in the delivered file so the client knows it is not yet an industry standard.
Audit a termbase for outdated or deprecated entries
23/30โจ What it does
Flags outdated termbase entries tied to product, branding, or industry changes and recommends an action per entry.
You are a terminology maintainer who periodically audits termbases for entries that no longer reflect current usage. <context> A client's termbase has not been reviewed in a while and I want to check which entries are likely outdated because the product, industry, or branding has moved on. </context> <inputs> - Termbase entries: [PASTE TERMBASE ENTRIES] - What has changed since the termbase was built: [DESCRIBE CHANGES, E.G. REBRAND, NEW PRODUCT NAME] - Target language: [TARGET LANGUAGE] </inputs> <task> Go through the termbase entries and flag any that reference the described changes, meaning old product names, deprecated features, or outdated branding, and note whether the entry should be updated, removed, or kept as a historical reference. </task> <constraints> Do not flag an entry as outdated unless it clearly connects to the described changes, do not guess at unrelated issues. Recommend a specific action per entry, not a vague note. Keep entries that are still valid out of the flagged list entirely. </constraints> <format> Return a table with columns: Term, Reason flagged, Recommended action (update, remove, keep as historical). </format>
Pro tip: Schedule this audit every time a client tells you about a rebrand or major product change, not on a fixed calendar, the trigger is the change itself.
Standardize terminology across translators on the same project
24/30โจ What it does
Standardizes conflicting terminology across sections handled by different translators on the same project.
You are a lead translator coordinating terminology consistency across a team working on the same large project. <context> Multiple translators are working on different sections of the same large document and I need to standardize the terms before we combine sections, since each translator has worked somewhat independently. </context> <inputs> - Section 1 key terms and translations used: [PASTE SECTION 1 TERMS] - Section 2 key terms and translations used: [PASTE SECTION 2 TERMS] - Section 3 key terms and translations used, if any: [PASTE SECTION 3 TERMS OR WRITE NONE] - Target language: [TARGET LANGUAGE] </inputs> <task> Compare the terms used across sections, identify every case where the same source concept was translated differently by different translators, and propose one standardized term for each with a brief reason. </task> <constraints> Pick the standardized version based on accuracy and consistency with the domain, not simply the first one used. Note if none of the existing versions are ideal and propose a better one instead. Keep the output actionable so it can be sent straight to the team as a terminology decision list. </constraints> <format> Return a table with columns: Source concept, Version used in each section, Standardized decision, Reason. End with instructions for the team on how to apply the find-and-replace. </format>
Pro tip: Share the decision list with the whole team before the final merge, not after, so nobody has to redo completed sections.
Generate a terminology change log entry for client records
25/30โจ What it does
Documents a client-approved terminology change as a clean, appendable change log entry.
You are a terminology manager who keeps a formal change log for client-approved terminology decisions. <context> A client just approved a change to an established term and I need to log it properly so the whole team and future translators know the term changed and why. </context> <inputs> - Old term and its translation: [OLD TERM AND TRANSLATION] - New approved translation: [NEW APPROVED TRANSLATION] - Reason for the change: [REASON FOR CHANGE] - Date and who approved it: [DATE AND APPROVER] </inputs> <task> Write a clear change log entry documenting the old term, the new term, the reason for the change, who approved it, and what existing content, if any, should be updated to reflect the new term. </task> <constraints> Keep the entry factual and short, this is a record, not a narrative. Explicitly flag whether existing published content needs retroactive updates or whether the change applies going forward only. Use a consistent format so future entries can be appended to the same log. </constraints> <format> Return a single change log entry with fields: Date, Old term, New term, Reason, Approved by, Applies to existing content (yes or no, with scope if yes). </format>
Pro tip: Keep every entry in one running document per client so a new translator can read the full terminology history in one place.
Most people use 10% of Claude. Tutorials unlock the rest.
AI Academy: 300+ hands-on tutorials on Claude, ChatGPT, Midjourney, and 50+ AI tools. New tutorials added every week.
Client and Project Communication
5 promptsWrite a scope clarification email for an ambiguous translation request
26/30โจ What it does
Drafts a short, professional email requesting the missing details needed to scope and quote a translation job.
You are a freelance translator who writes clear scope clarification emails before quoting a job. <context> A client sent me a translation request that is missing key details I need before I can quote or start the work. </context> <inputs> - Client's original request: [PASTE CLIENT REQUEST] - What is missing or unclear: [LIST WHAT IS MISSING] - Target language: [TARGET LANGUAGE] </inputs> <task> Write a short, professional email that asks the client for the missing details, such as word count, deadline, intended use of the text, and any glossary or reference material, without sounding like I am creating obstacles. </task> <constraints> Keep the email under 150 words. Ask no more than four questions, prioritized by what most affects the quote. Keep the tone collaborative, not bureaucratic. </constraints> <format> Return the email with a subject line and a short signoff placeholder [YOUR NAME]. </format>
Pro tip: Always ask about intended use of the text, that one answer often changes the register and price more than any other detail.
Draft a rate and turnaround explanation for a rush job
27/30โจ What it does
Writes a confident, factual message explaining rush pricing and realistic turnaround to a client.
You are a professional translator explaining rush pricing to a client in a clear, non-defensive way. <context> A client asked for a fast turnaround on a translation and I need to explain the rush rate and realistic timeline without sounding like I am penalizing them. </context> <inputs> - Word count: [WORD COUNT] - Requested deadline: [REQUESTED DEADLINE] - Standard turnaround for this word count: [STANDARD TURNAROUND] - Rush rate or policy: [RUSH RATE OR POLICY] </inputs> <task> Write a short message that explains why the requested deadline requires a rush rate, states the realistic delivery time, and offers one alternative such as a phased delivery if that helps meet part of the deadline. </task> <constraints> Keep it factual and confident, not apologetic. State the numbers plainly. Offer the phased delivery option only if it is realistic given the word count and deadline, otherwise omit it. </constraints> <format> Return a short message, three to five sentences, ready to send. </format>
Pro tip: Always state the standard turnaround alongside the rush rate, the contrast is what makes the rush fee feel justified rather than arbitrary.
Summarize a long translator-client email thread into an action list
28/30โจ What it does
Extracts agreed decisions, open questions, and scope changes from a long client email thread.
You are a project coordinator who summarizes long client email threads into clear action items. <context> A translation project has generated a long back-and-forth email thread with the client and I need a clean summary of what was actually agreed before I proceed. </context> <inputs> - Full email thread: [PASTE EMAIL THREAD] - Project name: [PROJECT NAME] - Current date: [CURRENT DATE] </inputs> <task> Read the thread and extract what was actually agreed on, what is still undecided, and any deadline or scope changes mentioned along the way. </task> <constraints> Do not include pleasantries or back-and-forth that did not change a decision. If two messages in the thread contradict each other, flag the contradiction rather than picking one silently. Order the action items by what needs to happen first. </constraints> <format> Return three sections: Agreed decisions, Still undecided (with the open question), Deadline or scope changes noted. </format>
Pro tip: Send this summary back to the client as a quick confirmation email, it catches misunderstandings before they become delivery problems.
Write a polite pushback on an unreasonable revision request
29/30โจ What it does
Drafts a professional, solution-oriented response to a client revision request that exceeds the original scope.
You are a professional translator who responds to client revision requests that go beyond the original scope or brief. <context> A client has asked for a revision that goes beyond correcting an error, it is actually a new stylistic preference or scope change, and I need to respond professionally without damaging the relationship. </context> <inputs> - Client's revision request: [PASTE REVISION REQUEST] - Original brief or agreed scope: [PASTE ORIGINAL BRIEF] - What I am willing to do: [WHAT YOU ARE WILLING TO DO, E.G. ONE FREE PASS, OR ADDITIONAL FEE] </inputs> <task> Write a response that acknowledges the client's request, explains clearly why it falls outside the original scope, and proposes what I am willing to do, whether that is a paid additional round or a boundary on future scope. </task> <constraints> Do not sound defensive or accusatory. Reference the original brief specifically rather than making a general statement about scope. Keep the tone solution-oriented, ending on the proposed path forward, not on the pushback itself. </constraints> <format> Return a short email, four to six sentences, with a clear next step at the end. </format>
Pro tip: Always quote the specific line from the original brief, a vague scope objection is much easier for a client to dispute than a specific one.
Create a project handoff note for another translator covering your work
30/30โจ What it does
Produces a scannable handoff note covering status, open items, and terminology notes for a colleague taking over a project.
You are a translator preparing a handoff note for a colleague who will continue your work while you are unavailable. <context> I need to hand off an in-progress translation project to another translator for a period of time and I want them to pick it up without losing context or introducing inconsistencies. </context> <inputs> - Project name and client: [PROJECT NAME AND CLIENT] - Work completed so far: [DESCRIBE WORK COMPLETED] - Open decisions or pending client questions: [LIST OPEN ITEMS] - Key terminology or style notes not yet documented elsewhere: [KEY NOTES OR WRITE NONE] </inputs> <task> Write a handoff note that gives the covering translator everything they need: current status, what is left to do, any pending client questions, and terminology or style notes that are not written down anywhere else yet. </task> <constraints> Keep it scannable, this will be read quickly by someone stepping in cold. Flag anything time-sensitive at the top, such as an approaching deadline or an unanswered client question blocking progress. Do not bury the most urgent item in the middle of the note. </constraints> <format> Return sections in this order: Urgent or time-sensitive items, Current status, Remaining work, Terminology and style notes, Client contact info placeholder [CLIENT CONTACT]. </format>
Pro tip: Put the urgent items section first every time, a covering translator reading top to bottom should hit the deadline risk before anything else.
Free tool
Prompt Optimizer
Turn a rough idea into a structured, professional AI prompt.
Frequently Asked Questions
Prompts are the starting line. Tutorials are the finish.
A growing library of 300+ hands-on tutorials on ChatGPT, Claude, Midjourney, and 50+ AI tools. New tutorials added every week.
7-day free trial. Cancel anytime.
Related guides