36 Best Mistral Le Chat Prompts (2026)

You are my executive assistant. Here are the emails from my inbox: [PASTE OR CONNECT GMAIL]. Triage them into: Reply now, Reply later, Delegate, and Ignore. For each "Reply now", draft a reply under 80 words in my voice (direct, warm, no filler). Output a table with columns Sender, Why it matters, Suggested action, and put the drafts below it.

Turn this meeting transcript into structured notes: [PASTE TRANSCRIPT]. Output three sections: Decisions made, Action items (as a table with Owner, Task, Due date), and Open questions. Do not invent details that are not in the transcript, and flag anything left unresolved. Keep it tight.

Here is everything on my plate this week: [PASTE LIST]. Act as a sharp chief of staff. Sort it by impact and urgency, name the three things that actually move the needle, and say what to delegate or drop. Lay it out as a simple day-by-day plan, and call out anything that looks like busywork.

Read the attached document and give me: a 3-sentence executive summary, the 5 most important points as bullets, any numbers or dates that matter, and one section titled "What I should do about this." If the document contradicts itself or leaves something unclear, say so. Keep the whole thing under 300 words.

You are a delivery lead. Turn this rough idea into a one-page project brief: [DESCRIBE PROJECT]. Cover: the goal in one sentence, scope (in and out), key milestones with rough dates, the main risks, and who needs to be involved. Use Work mode to think it through. Output clear sections, no filler, ready to paste into a doc.

Open a Canvas and draft a [10]-slide deck outline for [PRESENTATION TOPIC] aimed at [AUDIENCE]. For each slide give a title, three bullet points, and a one-line speaker note. Keep one idea per slide and lead with the takeaway. End with a slide that states the single ask. Let me edit it in place.

You are a writer for [AUDIENCE]. Write a [1,200]-word blog post titled "[TITLE]" from this outline: [PASTE OUTLINE]. Keep the tone practical and direct, lead each section with the takeaway, use short paragraphs and concrete examples, and avoid clichés and filler. Output in clean Markdown with H2 sections and a two-line intro.

Write a LinkedIn post for [MY ROLE] about [TOPIC / LESSON]. Open with a one-line hook that earns the scroll without clickbait, tell it as a short story or a concrete example, and end with a question that invites comments. Keep it under 150 words, use line breaks for readability, and no more than one emoji. Give me two hook options.

Rewrite the text below to be clearer and more direct without losing meaning. Cut filler, fix weak verbs, and keep my key terms. Target a [confident but plain] tone at a [general professional] reading level. Return the rewrite first, then a short bullet list of the main changes you made. Text: [PASTE TEXT]

You are writing a cold outreach email from me ([MY ROLE] at [MY COMPANY]) to [PROSPECT ROLE] at [PROSPECT COMPANY]. Goal: [GOAL]. Keep it under 90 words, open with a specific reason I am reaching out to them (not a template), state one clear value point, and end with a low-friction ask. Give me three subject-line options.

Turn these raw notes into a [500]-word newsletter issue for [AUDIENCE]: [PASTE NOTES]. Structure: a strong one-line opener, 3-4 short sections with subheads, and a closing line with one clear call to action. Keep my voice casual and specific, and suggest a subject line and preview text. Output in Markdown.

Generate 12 headline options for [PIECE / PRODUCT] aimed at [AUDIENCE]. Mix angles: benefit-led, curiosity, contrarian, number-led, and how-to. Keep each under 65 characters, avoid hype words, and label the angle of each. Then pick your top three and say why they would perform best.

I am getting this error in [LANGUAGE / FRAMEWORK version]: [PASTE ERROR]. Here is the relevant code: [PASTE CODE]. Think it through step by step, find the root cause, give me the corrected code, and explain what was wrong in two sentences. If more than one cause is possible, rank them by likelihood.

Explain what this code does to a developer new to [LANGUAGE / LIBRARY]: [PASTE CODE]. Give a one-paragraph overview, then a block-by-block walkthrough of the tricky parts, and call out any bugs, edge cases, or performance concerns you notice. Keep it concrete.

Write thorough unit tests for this function using [TEST FRAMEWORK]: [PASTE FUNCTION]. Cover the happy path, edge cases, invalid inputs, and boundary conditions. Explain in one line what each test verifies, and note any behavior in the function that looks like a bug or is ambiguous enough to need a spec decision.

Write a [PostgreSQL] query for this request: [WHAT I WANT]. Here is the schema: [PASTE TABLES / COLUMNS]. Return the query, a one-line explanation of each join or filter, and a note on indexes that would keep it fast at scale. If my request is ambiguous, state the assumption you made.

Refactor this code for readability and maintainability without changing behavior: [PASTE CODE]. Improve naming, reduce nesting, and extract logic where it helps. Return the refactored version, then a short bullet list of what you changed and why. Keep it idiomatic for [LANGUAGE] and do not add dependencies.

I want to hand this off to a Vibe cloud agent. Write a precise task spec it can run end to end in an isolated sandbox: [DESCRIBE THE TASK, e.g. fix the failing tests in module X]. Include the exact acceptance criteria, the files or areas it may touch, what it must NOT change, and how it should report back. Keep it unambiguous.

Search the web and give me a briefing on [TOPIC] as of this month. Cover: the current state, the three biggest developments in the last six months, the key players, and the open questions. Cite every claim with a link, prefer primary and recent sources, and note the publication date next to each fact. Output a bulleted brief under 400 words, then a one-line bottom line.

Search the web and compare [PRODUCT A] and [PRODUCT B] for a [USE CASE] buyer. Build a table: price, key features, real user complaints, and best-fit user. Pull from current reviews and forums, cite every source with a link and date, and end with a clear recommendation for someone who cares most about [PRIORITY].

Fact-check this claim using web search: "[CLAIM]". Find the original source, what the primary evidence actually says, and how credible outlets have covered it. Rate it True / Misleading / False / Unproven with a short reason. Cite every source with a link and date, and separate what is confirmed from what is still uncertain.

Use Work mode and web search to map the current landscape of [MARKET / INDUSTRY]. Cover the main players and their positioning, recent funding or M&A, pricing norms, and where the market is heading. Cite every source with a link and date, and end with two gaps or openings a new entrant could target.

Read this URL and summarize it for me: [PASTE URL]. Give a 3-sentence overview, the key points as bullets, any data or claims worth double-checking, and one line on what the author wants me to do or believe. If the page is thin or promotional, say so. Keep it under 200 words.

Use Work mode and web search to review the current evidence on [RESEARCH QUESTION] published in the last two years. Summarize what the strongest sources agree on, where they disagree, and what remains unsettled. Cite each source with a link and date, weight studies over opinion pieces, and end with a short "confidence and gaps" note.

Here is a CSV: [UPLOAD FILE]. Using the code interpreter, load it, describe the columns and row count, and answer these questions: [YOUR QUESTIONS]. Show the key numbers, flag any data-quality issues (missing values, outliers, duplicates), and end with the three findings that matter most for [DECISION]. Do not speculate beyond the data.

I need a [Google Sheets] formula/model to [GOAL]. My data is laid out like this: [DESCRIBE COLUMNS]. Give me the formula, explain what each part does, and note common ways it could break (blank cells, text vs numbers). If a small helper column would make it cleaner, suggest it.

Extract structured data from the text below into valid JSON only — no prose. Use exactly this schema: {"company": string, "contact": {"name": string, "email": string}, "items": [{"sku": string, "qty": number, "price": number}], "total": number}. If a field is missing, use null. Do not invent values. Text: [PASTE TEXT]

Using the code interpreter, make a clean chart from this data: [PASTE OR UPLOAD]. Chart type: [bar / line / scatter]. Put [X] on the x-axis and [Y] on the y-axis, label the axes and title clearly, and use a colorblind-friendly palette. Then tell me in two sentences what the chart shows and one thing that stands out.

Read the attached [report / whitepaper] and produce: a one-paragraph executive summary, the key findings as bullets with the page numbers, any figures or tables worth quoting, and a "so what for [MY ROLE]" section. Note anything the report claims without evidence. Keep the whole thing to one page.

Here is a scanned [invoice / receipt / form]: [UPLOAD IMAGE OR PDF]. Read the text (OCR), then extract the fields into a table with columns [FIELD 1], [FIELD 2], [FIELD 3]. Preserve exact numbers and dates, and mark any field you could not read clearly as "unclear" rather than guessing. Output the table only.

Translate the text below from [SOURCE LANGUAGE] into [TARGET LANGUAGE] for a [AUDIENCE / REGION]. Do not translate word-for-word — localize idioms, units, currency, and tone so it reads as if written by a native speaker. Keep the meaning and any names or product terms exact. Return the translation, then a short note on any choice you had to make. Text: [PASTE TEXT]

Write a [formal business email] in [French / German / Spanish / Italian] from [MY ROLE] to [RECIPIENT] about [PURPOSE]. Match the professional register and courtesy conventions native to that language and country, not a translated English tone. Keep it under [120] words, end with one clear ask, and add an English gloss below so I can check it.

Help me build a reusable agent in Le Chat's Agent Builder that [WHAT IT DOES, e.g. summarizes new support tickets and drafts a reply]. Write: the agent's system instructions, the inputs it should expect, the tools it needs (web search, code, connectors), the output format, and 3 test inputs with the expected output. Keep the instructions specific enough that it behaves the same every run.

Generate an image: [SUBJECT with 2-3 defining details], in [ENVIRONMENT], lit by [LIGHTING], in a [SINGLE STYLE, e.g. editorial product photography]. Shot [CAMERA / LENS or ANGLE], [MOOD]. Aspect ratio [16:9 / 1:1 / 9:16]. Keep one coherent style and skip quality-adjective spam like "8k ultra masterpiece."

You are a pragmatic strategy advisor for a [COMPANY TYPE]. We are deciding whether to [DECISION]. Context: [KEY FACTS AND CONSTRAINTS]. Use Work mode to reason it through, then lay out the options, the main trade-offs, the biggest risk, and what you would do and why. Be blunt about weaknesses. Output a one-page brief: recommendation, rationale, risks, next step.

You are a classification function. For each input below, return valid JSON only matching this schema: {"category": one of ["billing","technical","account","other"], "priority": one of ["low","medium","high"], "summary": string (max 15 words)}. No prose, no markdown fences. If the input is empty, return {"category":"other","priority":"low","summary":"empty input"}. Inputs: [PASTE ITEMS]