Support

help --bug-report

back to chat

Report a Bug

Found an issue? Help us improve by reporting it! We read every report.

Click to upload or drag & drop
PNG, JPG, GIF, WebP (Max 5MB)

Getting Started

Creating Your Account

  1. On the login page, click $ register
  2. Fill in your username, email and password
  3. Click $ submit registration
  4. Check your email inbox (and spam folder) for a verification link
  5. Click the verification link to activate your account
  6. Return to the login page and sign in

Starting a Conversation

  1. After logging in, click + new chat in the left sidebar
  2. Type your message in the input box at the bottom
  3. Press Enter or click the send button to get a response
  4. The AI will stream its response in real-time
Your conversations are saved automatically. You can return to any previous chat by clicking on it in the sidebar. Chats can be renamed, archived, or deleted.

About exploit.bot AI

exploitbot-v7 is our custom uncensored AI model, purpose-built for offensive security and development:

  • Exploit Development - Write exploits, payloads, shellcode, and PoC code
  • Penetration Testing - Guidance on recon, enumeration, privilege escalation, post-exploitation
  • Malware Analysis - Analyze and understand malware behavior, write detection rules
  • Security Tool Creation - Build scanners, fuzzers, RATs, C2 frameworks, and custom tools
  • Coding & Development - Write, debug, and optimize code in any language
  • CTF Challenges - Help with capture-the-flag competitions across all categories
Example prompts:
  • "Write a Python port scanner with banner grabbing"
  • "Create a reverse shell payload for Linux"
  • "Explain how SQL injection works and write a tool to test for it"
  • "Help me solve this CTF challenge: [description]"
  • "Build a keylogger in C++ for Windows"

Chat Controls

These controls appear in the chat composer area:

Web Search on by default

Click the magnifying glass 🔍 to toggle web search on or off.

  • Enabled by default - searches the internet automatically when relevant
  • Shows what it's searching for and the results found
  • Finds current CVEs, documentation, GitHub repos, security tools
  • Auto-detects security/code queries and searches GitHub specifically
  • Search results are used by the AI to write accurate, up-to-date code
  • Subject to plan limits (free: 3 searches per 5 hours)

ExploitDump Breach Check on by default

Automatically activated when you ask about breaches, leaks, or compromised credentials.

  • Check emails for known data breaches (sources, masked usernames, dates)
  • Check domains for employee/customer/third-party breach counts + individual entries
  • Check passwords against breach databases (k-anonymity, safe)
  • Results appear automatically in the chat — just ask naturally
  • Subject to plan limits (free: 1 breach check per 5 hours)

Extended Thinking optional

Toggle the Think button in the chat composer to enable or disable reasoning.

  • Off (default): Quick, direct answers — faster responses
  • On: Deep reasoning — the AI shows its full thought process in a collapsible "Reasoning" box above the response

Enable thinking for complex exploits, architecture decisions, or multi-step problems.

File Upload optional

Click the 📎 paperclip button to attach a text or code file to your message.

  • Maximum file size: 50KB
  • Supports text and code files only (no images or binaries)
  • File content is included in your message for the AI to analyze

Artifacts System

Code blocks from AI responses are automatically extracted as artifacts.

Artifacts Sidebar

  • Click { } in the header to open the artifacts sidebar
  • Shows all code blocks from the current conversation
  • Each artifact shows filename, language, and line count
  • Click an artifact to preview it with syntax highlighting
  • Filenames are auto-inferred from code content (class names, shebangs, etc.)

Artifact Actions

  • Copy: Copy code to clipboard
  • Download: Download as a file with the correct extension
  • Save: Manually save to your persistent artifacts collection
  • Download All: Download every artifact from the current chat

Auto-Save & Persistent Storage

  • Artifacts are automatically saved to your account when generated
  • View all saved artifacts at artifacts page
  • Upload your own files (code, images, documents) to the artifacts page
  • Maximum 200 artifacts per user (oldest auto-deleted when limit reached, pinned files excluded)
  • Edit, rename, download, or delete artifacts from the artifacts page

Chat Features

Code Blocks

Code in AI responses is formatted with syntax highlighting. Each code block has:

  • Language label: Shows the detected programming language
  • Copy button: One-click copy to clipboard
  • Download button: Download the code as a file

Reasoning Box

When extended thinking is enabled, the AI's thought process appears in a collapsible Reasoning section above the response. Click to expand/collapse.

Chat Management

  • New Chat: Click + new chat to start a fresh conversation
  • Rename: Click on a chat title in the sidebar to rename it
  • Archive: Archive old chats to keep the sidebar clean
  • Delete: Permanently remove chats you no longer need
  • Cancel: Stop the AI mid-response by clicking the stop button

Projects

  • Organize related chats and code files into projects
  • Set custom system prompts per project
  • Link artifacts and chats to a project
  • Access from projects in the sidebar

Understanding Tokens

What Are Tokens?

Tokens are the "units" that AI models use to measure text. Think of them as pieces of words:

  • Short common words = 1 token (e.g., "the", "is", "a")
  • Longer words = 2-3 tokens (e.g., "programming" = 2 tokens)
  • Code and special characters may use more tokens
Rough estimates:
  • 1 token ~ 4 characters or 0.75 words
  • 100 tokens ~ 75 words
  • 1,000 tokens ~ 750 words (about 1.5 pages)

How Token Usage Works

  • Input tokens: Your message + conversation history sent to the AI
  • Output tokens: The AI's response
  • Total usage: Input + Output tokens counted against your plan
  • Rolling window: Usage resets on a rolling 5-hour basis
Longer conversations use more tokens because the AI reads the entire chat history. Start a new chat when switching topics to keep token usage low.

Usage Plans

Usage is tracked over a rolling 5-hour window. When your limit resets, you can continue chatting.

Plan Tokens / 5h Per Chat Searches Breach Checks
Free 15,000 8,000 3 1
Beta Testers 50,000 80,000 105 3
Pro 100,000 100,000 20 10
Supporter 200,000 Unlimited Unlimited 20
Enterprise 500,000 200,000 Unlimited Unlimited
Unlimited No limit Unlimited Unlimited Unlimited

What Happens When You Hit a Limit?

  • Token limit: You'll see a message telling you how long until your limit resets
  • Per-chat limit: Start a new chat to continue (this keeps conversations focused)
  • Search limit: The AI will respond using its built-in knowledge instead of searching the web

Contact [email protected] for plan upgrades.

Tips for Best Results

Writing Good Prompts

  • Be specific: "Write a Python reverse shell with encryption using AES" is better than "help with hacking"
  • Provide context: Tell the AI what OS, language, and goal you have
  • Include code: Paste relevant code if you need help debugging or extending it
  • Specify constraints: Language, framework, target OS, evasion requirements

When to Use Each Feature

  • Web Search ON: Finding CVEs, latest tools, GitHub repos, current documentation
  • Web Search OFF: General coding, creative tasks, or when you want pure AI knowledge
  • Thinking ON: Complex exploit chains, multi-step attacks, architecture decisions
  • New Chat: When switching topics (saves tokens and keeps context clean)

Troubleshooting

  • AI seems confused: Start a new chat - long conversations lose context
  • Response cut off: Ask the AI to "continue" or "finish the code"
  • Wrong language/framework: Specify exactly what you want in your prompt
  • Token limit reached: Wait for the 5-hour window to reset, or start a new chat if you hit the per-chat limit
  • Search not working: Make sure the 🔍 toggle is active (highlighted). You may have hit your search limit.

Need More Help?

Contact the admin team:

Join our community:

Last updated: February 2026