THE MEDIA FOR AI THAT ACTUALLY HELPS.English edition · Understand & take action
Spotted on the web

Extract the lines of a PDF command to power the commercial software, from a roofer

Use described publicly by France Num — Pierre Fruneau (Tri-Énergie) (6 January 2026), summarised by the desk. Not tested by iadujour.

The result to aim for

Pierre Fruneau, manager of Tri-Energy in Bouaye (coverage and photovoltaics), connected n8n to Outlook: the tool reads the mail and the PDF, creates the order in the commercial software, then stores the message as "treated by AI".

What was done

France Num published on 6 January 2026 the testimony (video U2P Pays de la Loire, with transcription) of Pierre Fruneau, managing since 2006 a cover TPE, Velux, insulation and photovoltaics. For the past two years, Fruneau claims to ear "about 23 hours a week"; the page notes that the number could also be "2 to 3 hours" depending on the speech rate. Franck Goron, technical-commercial, says he discovers the tool and does not have "a direct impact for now": separate sheet not contained. Limit 3/URL.

Before you start

Tool cited by the source: n8n. Fill in the bracketed items: [coller]. Strip confidential data that is not needed and check your tool’s access terms. Features and plans can change.

The method, in three steps.

  1. Gather the context

    Spot the bracketed items in the prompt ([coller]) and prepare what they ask for.

  2. Adapt the prompt

    Copy the prompt below, replace the bracketed items, then paste it into the AI assistant you use. The cited tool (n8n) is not necessarily available to the public..

  3. Reread before using

    Reread the result, correct it if needed and keep the final decision. Caveats are detailed below.

Your starting point.

Replace the bracketed items, then adapt the proposal to your situation. This prompt is written by the iadujour desk It is not a quotation from the source.

The prompt to customise
Here is the text of an e-mail provider and the content of an order PDF (without personal data): [coller].

1. List the lines (reference, designation, quantity, unit).
2. Report what is illegible or ambiguous.
3. Proposes a table ready to enter in the commercial software.
Do not invent any reference or quantity.

What deserves your attention

IA is not a shopkeeper: a poor reference in the PDF creates a bad order. Cross-check the extracted lines with the right supplier before purchasing, and do not paste customer data into an uncontrolled stream.
Origin of the case. Use described publicly by France Num — Pierre Fruneau (Tri-Énergie) (6 January 2026).
Proof level: Institutional source. Source checked on 11 November 2026. Use described by the source, not tested by the desk. Original summary written by iadujour.fr: figures and claims are the source’s; the prompt is a desk proposal, not a quotation.

Useful skills

Official skills on the same task, paired by the desk: they are not part of the described case.

  • Everything with PDFsAnthropic · Covers most operations on PDFs: extract text and tables, merge, cut, rotate, watermark, create, fill out forms, encrypt or decipher, extract images and make a scan searchable by OCR.
TO GO FURTHER

Other ideas to put to work.

The whole Admin theme
Admin

Prepare an association grant application file

Structure the project, budget and indicators based on the terms of reference of the funder and the actual data of the association.

Tool: Type: Practical case
Admin

Automatically turn an order email into a purchase order

An artisan covering the Loire-Atlantique has put in place a n8n stream with the AI which reads the orders received by e-mail, extracts the lines and creates them in its commercial software.

Tool: n8nType: Spotted
DISCUSSION

Comments.

A field note, a question, a precision on this use case? Comments are reviewed before publication.

Published comments

Loading comments…

    Leave a comment

    Moderation charter. Every comment is reviewed before publication. We refuse hate speech, insulting or discriminatory language, advertising, spam and promotional links, off-topic posts and other people’s personal data.