Prompts, Skills & Knowledge PacksAI Agents & Digital WorkersIndustry-Specific AIModels, APIs & AI InfrastructureDownload

Translated automatically.

KEO-UDVI-EBP: Master Prompt for the Medical-Social AI Framework

This listing has not been audited by AIMarket24. We haven't verified that the code or workflow works as described. Payment protection still applies.

An open-source framework and strict safeguards to secure the use of LLMs in supporting vulnerable individuals.

KEO-UDVI-EBP : Master Prompt pour l'Encadrement de l'IA Médico-Sociale

About this app

KEO-UDVI-EBP is an operational prompt framework (master prompt) designed for AI assistants used in high-stakes contexts: mental health, social support, crisis management, and career guidance. This system acts as a safety layer. Instead of allowing the AI to improvise, it forces the model to follow a strict methodology with explicit safeguards: prohibition on making medical diagnoses, prohibition on prescribing, obligation to direct to local emergency services in case of danger, and a strict rule prohibiting the transposition of an emergency number from one country to another. It is designed for developers, associations, and professionals who wish to deploy AI in sensitive environments while needing a documented, auditable, and portable safety framework.

Problem it solves

Deploying a generalist LLM in a medical-social context without safeguards is risky: the AI can hallucinate, make a diagnosis, prescribe medication, or provide an incorrect emergency number. KEO-UDVI-EBP solves this problem by providing a ready-to-use prompt framework that imposes epistemic rigor (distinction between facts and hypotheses), territorial safety (never mixing emergency resources from different countries), and non-substitution (the AI cannot replace a human professional). It offers developers a documented safety base without having to write hundreds of pages of safety instructions.

Key features

  • ✦ Portable core
  • ✦ 20 operational modes
  • ✦ 7 levels of reasoning (K0-K7)
  • ✦ 47 frameworks
  • ✦ 90 guidance sheets (30 PSY + 60 MS)
  • ✦ routing index P001-P960
  • ✦ protection against prompt injection
  • ✦ territorial resource rules
  • ✦ crisis protocol with direction to local emergency services
  • ✦ anti-hallucination epistemic labeling

Tech & operating

Built withClaudeGPTGemini
Tech stackMarkdownGitHub
Operating modelManaged / hybrid
External APIsNone
AI modelsGPT-4oClaude 3.5 SonnetGemini 1.5 ProMistral Large
Monthly running cost$0 / mo
Setup difficultyeasy
DocumentationIncluded

Ownership & legal

Proof of ownershipGitHub repository (av974/Keo-udvi-ebp) with commit history, Copyright01 official deposit certificate (Ref: CR-2026-37263) for the keo-udvi-ebp.md file.
AuthorshipNot disclosed
Open-source partsNo third-party code is included. Distributed under Creative Commons Attribution 4.0 International (CC BY 4.0).
Included assetsSource file (keo-udvi-ebp.md)Copyright certificateGitHub repository access
Transferred accounts-
Managed handoffYes

Known issues

The complete framework exceeds 120,000 characters, which exceeds the context window of many standard AI models. The 90 medical-social guidance sheets are non-substitutive guides and do not constitute clinical validation or certification. The framework does not include an executable tool or API; it relies on the tools and permissions of the host platform. No independent clinical validation has been conducted.

Report this listing

Tell us if this content is illegal or harmful. This notice goes to AIMarket24. It is not the code-audit report.

Questions before you buy

What do I get?

Prompts, Skills & Knowledge Packs. Download of the files after payment. Open source. Documentation included. Source file (keo-udvi-ebp.md). Copyright certificate.

What will I be charged?

Asking price $3,100. Stripe charges this amount. Nothing is added on top. The seller's commission comes out of this price, not from an extra buyer fee.

What if the file is not what the listing describes?

You pay on Stripe. AIMarket24 holds the payment until you confirm the delivery matches this listing. If you do not open a dispute, the order confirms automatically 3 days after delivery. You can open a dispute from the order page before that.

Was this listing audited?

This listing has not been audited by AIMarket24. We haven't verified that the code or workflow works as described. Payment protection still applies.

Related listings

AI readability: make a site readable by AI, with a pass test
Prompts, Skills & Knowledge PacksClaudeGPT
Buy now
$10
Marketing, SEO & Growth
Bank statement PDF to Excel (Claude skill)
Prompts, Skills & Knowledge PacksClaudePython
Buy now
$15
Data, Analytics & BI
Spreadsheet cleanup for messy tables (Claude skill)
Prompts, Skills & Knowledge PacksClaudePython
Buy now
$9
Data, Analytics & BI