# prompt-injection-detector

> Detects hidden prompt injection instructions that might manipulate AI models like Copilot and Claude.

Canonical page: [https://chrome-stats.com/d/prompt-injection-detector](https://chrome-stats.com/d/prompt-injection-detector)

## Overview

- **ID:** `prompt-injection-detector`
- **Platform:** Firefox
- **Type:** Firefox add-on
- **Status:** Available
- **Publisher:** Vishesh Agarwal
- **Category:** privacy-security
- **Version:** 1.0.0
- **Last updated:** 2026-04-07
- **First published:** 2026-04-04
- **Size:** 20 KB
- **Data as of:** 2026-09-06
- **Store listing:** [Firefox Add-ons Store](https://addons.mozilla.org/firefox/addon/prompt-injection-detector/)
- **Privacy policy:** [https://addons.mozilla.org/firefox/addon/prompt-injection-detector/privacy/](https://addons.mozilla.org/firefox/addon/prompt-injection-detector/privacy/)

## Description

AI assistants like GitHub Copilot, ChatGPT, and others read web page content when you ask them to help. Attackers can hide malicious instructions in that content — invisible to you, but visible to the AI — to hijack its behaviour, steal your data, or bypass safety filters.

<strong>PromptGuard detects:</strong>
- Hidden elements (<code>display:none</code>, <code>visibility:hidden</code>, zero opacity, sub-pixel fonts, same-colour text)
- HTML comments — invisible to humans but read by AI tools ingesting page source
- LLM-specific formats: <code>[INST]</code>, <code>system:</code>, <code>assistant:</code> prompt injection patterns

<strong>Three sensitivity levels:</strong>
- 🟢 Normal — high-confidence imperative overrides only (low false positives)
- 🟠 High — adds jailbreak, DAN, developer-mode, bypass patterns
- 🔴 Ultra — adds roleplay, persona, exfiltration, and LLM prompt-format patterns

Click any finding to flash and scroll to the exact element on the page.

All scanning runs locally in your browser. Nothing is sent anywhere.

## Rankings

- #143,484 — Overall
- #110 — inject

## Permissions and access

### Permissions

- `activeTab`
- `scripting`

### Host permissions

- `http://*/*`
- `https://*/*`

## Safety

- **Risk impact:** Moderate risk impact
- **Risk likelihood:** Moderate risk likelihood

### Analysis details

- Allows access to all websites, posing a significant security risk as it can monitor and modify data from any visited site.
- This extension has low user count. Unpopular extensions may not be stable or safe.

> Some risk analysis details are omitted from this free response. [Upgrade to view the full analysis](https://chrome-stats.com/pricing).

## Similar extensions and apps

- [PromptVigil](https://chrome-stats.com/d/promptvigil)
- [ContextHound](https://chrome-stats.com/d/contexthound) — 1 users
- [Prompt Leak Guard](https://chrome-stats.com/d/prompt-leak-guard)
- [Prompt Guard](https://chrome-stats.com/d/prompt-guard) — 1 users
- [AI Prompt Linter](https://chrome-stats.com/d/ai_prompt_linter) — 1 users
- [TeamPrompt AI Prompt Manager](https://chrome-stats.com/d/teamprompt-ai-prompt-manager) — 1 users
- [AI Prompt Injector](https://chrome-stats.com/d/ai-prompt-injector) — 6 users, 5.00 / 5
- [PromptProtect](https://chrome-stats.com/d/promptprotect)
- [CleanMyPrompt - AI Privacy Firewall](https://chrome-stats.com/d/cleanmyprompt-ai-firewall)
- [Prompt Armour — AI Privacy Guard](https://chrome-stats.com/d/prompt-armour-ai-privacy-guard) — 1 users
- [Prompt Injector for ai tools](https://chrome-stats.com/d/prompt-injector-for-ai-tools) — 6 users
- [PrecisePrompt](https://chrome-stats.com/d/preciseprompt)

---

Source: [Chrome-Stats](https://chrome-stats.com/d/prompt-injection-detector)