PromptShield AI: Zero-Leak Local Redaction & Prompt Hardening Engine for Open-Weight Models

Description and use

<p><em>This is a submission for the <a href="https://dev.to/challenges/hacktoberfest-week3-2026-10-19">Hacktoberfest Open-Source AI Challenge: Week 3</a>.</em></p> <h2> What I Built </h2> <p><strong>PromptShield AI</strong> is an open-source, local-first redaction proxy and prompt-hardening engine built to prevent sensitive credentials, client PII, and infrastructure secrets from leaking into open-weight AI models.</p> <p>As developers, freelancers, and enterprises adopt open-weight models like <strong>Google Gemma 2</strong>, <strong>Meta Llama 3.2</strong>, and <strong>Mistral</strong>, conf

Problem

<p><em>This is a submission for the <a href="https://dev.to/challenges/hacktoberfest-week3-2026-10-19">Hacktoberfest Open-Source AI Challenge: Week 3</a>.</em></p> <h2> What I Built </h2> <p><strong>PromptShield AI</strong> is an open-source, local-first redaction proxy and prompt-hardening engine built to prevent sensitive credentials, client PII, and infrastructure secrets from leaking into open-weight AI models.</p> <p>As developers, freelancers, and enterprises adopt open-weight models like <strong>Google Gemma 2</strong>, <strong>Meta Llama 3.2</strong>, and <strong>Mistral</strong>, conf

Outcome

<p><em>This is a submission for the <a href="https://dev.to/challenges/hacktoberfest-week3-2026-10-19">Hacktoberfest Open-Source AI Challenge: Week 3</a>.</em></p> <h2> What I Built </h2> <p><strong>PromptShield AI</strong> is an open-source, local-first redaction proxy and prompt-hardening engine built to prevent sensitive credentials, client PII, and infrastructure secrets from leaking into open-weight AI models.</p> <p>As developers, freelancers, and enterprises adopt open-weight models like <strong>Google Gemma 2</strong>, <strong>Meta Llama 3.2</strong>, and <strong>Mistral</strong>, conf

Evidence

PromptShield AI: Zero-Leak Local Redaction & Prompt Hardening Engine for Open-Weight Models

Source