💰 BUG BOUNTY HUNTER FREE Part of the Bug Bounty 60-Day Course Oauth Token Leakage - Day 32 of 60 · 53% complete ⚠️ Authorised targets and responsible disclosure only. OAuth testing means intercepting your own authorization flow on a programme-approved target. Redirect_uri testing is done against your own account using a domain you own or an authorised lab environment. Never intercept another user's token or capture authorization codes from real user sessions. Whenever I assess an application that offers…
Saturday, November 7, 2026
Wednesday, November 4, 2026
CORS Misconfiguration Bug Bounty 2026 — Find, Test & Report Cross-Origin Bugs | Bug Bounty Day 31 of 60
💰 BUG BOUNTY HUNTER FREE Part of the Bug Bounty 60-Day Course CORS misconfiguration - Day 31 of 60 · 51% complete ⚠️ Authorised targets and responsible disclosure only. CORS testing consists of sending custom Origin headers to a target's own API endpoints — the same API your account is authorised to use. The JavaScript PoC is built in your own controlled environment, never deployed against real users. Keep testing within programme scope and disclose responsibly. One thing I've noticed…
Monday, November 2, 2026
AI Agent Hijacking Lab: A Document That Hijacks the Bot
TaskBot is a tool-using agent. Ask it to 'summarise this doc' and it reads the document — and then obeys any instructions it finds inside that document as if you had typed them yourself. The bot cannot tell your orders apart from text it was handed to read. Lab at a glance: Advanced · 130 XP · about 20 minutes · runs in your browser, no signup or setup. Try it yourself first; the full solution is behind a spoiler…
AI Tool-Output Injection Lab: Hostile Search Results Drive the Agent
ResearchBot searches the web, then feeds the results straight back into its own context to reason over them. Its guardrails check what you say — but not what the search results say. Poison a result and the agent follows it. Lab at a glance: Advanced · 145 XP · about 20 minutes · runs in your browser, no signup or setup. Try it yourself first; the full solution is behind a spoiler below. [se_lab_cta id="ai-tool-output-injection-1"] 📋 In this walkthroughWhat this…
AI Plan Injection Lab: Corrupt the Agent's Scratchpad
PlannerBot writes a plan to a scratchpad, then executes the steps — and it reads the scratchpad back on every loop for 'transparency'. If your input lands in that scratchpad, it is read as a planned step, and the agent will dutifully carry it out. Lab at a glance: Advanced · 145 XP · about 20 minutes · runs in your browser, no signup or setup. Try it yourself first; the full solution is behind a spoiler below. [se_lab_cta id="ai-plan-injection-1"]…
AI Confused Deputy Lab: Turn the Agent's Privileges Against Its User
FinanceBot can move money between any accounts — it holds system-level privileges as a reconciliation tool. When a low-privilege user chats with it, it performs the actions they ask for using its privileges, without checking whether the user is allowed to do them. Lab at a glance: Advanced · 145 XP · about 20 minutes · runs in your browser, no signup or setup. Try it yourself first; the full solution is behind a spoiler below. [se_lab_cta id="ai-confused-deputy-1"] 📋 In…
AI LLM-Driven SSRF Lab: Push the Agent's Fetch Tool Inward
FetchBot can retrieve URLs for you with a server-side tool. Its only guard is a denylist: it refuses URLs that contain 'localhost', '127.0.0.1' or 'internal'. Denylists miss things — and the cloud metadata endpoint does not contain any of those words. Lab at a glance: Advanced · 145 XP · about 20 minutes · runs in your browser, no signup or setup. Try it yourself first; the full solution is behind a spoiler below. [se_lab_cta id="ai-llm-ssrf-1"] 📋 In this walkthroughWhat…
AI Output XSS Lab: When the Chatbot's Reply Is Trusted HTML
ChatBotApp wanted ChatGPT-style rich formatting, so it renders the model's reply with innerHTML. But an LLM reply is just text an attacker can influence — and dropping untrusted text into innerHTML is how stored XSS has worked for twenty years. Lab at a glance: Advanced · 140 XP · about 20 minutes · runs in your browser, no signup or setup. Try it yourself first; the full solution is behind a spoiler below. [se_lab_cta id="ai-llm-output-xss-1"] 📋 In this walkthroughWhat this…
AI Token Forgery Lab: When the Backend Trusts a Model-Made Token
AuthBot's LLM writes a session token into its reply after you authenticate, and the backend trusts whatever token comes back from the model as real. But an LLM cannot keep a secret or sign anything — so a token it 'mints' is only as trustworthy as a sentence you can talk it into writing. Lab at a glance: Advanced · 145 XP · about 20 minutes · runs in your browser, no signup or setup. Try it yourself first; the…
AI Output Schema Bypass Lab: Valid JSON, Impossible Values
ApprovalBot reviews expense reports and returns a JSON verdict the backend validates against a schema: right types, required fields present. But a schema checks shape, not sense — so a verdict approving a $1,000,000 expense is perfectly valid JSON, and perfectly against policy. Lab at a glance: Advanced · 145 XP · about 20 minutes · runs in your browser, no signup or setup. Try it yourself first; the full solution is behind a spoiler below. [se_lab_cta id="ai-output-schema-bypass-1"] 📋 In…
AI Agent State Lab: Tamper the Saved Role to Reach Admin Code
StateBot saves its agent state as a JSON blob in your browser storage, and on every request it reloads that blob and reads a role field to decide what it can do. State you can edit is state you control — so set the role to admin and the admin code path opens. Lab at a glance: Advanced · 145 XP · about 20 minutes · runs in your browser, no signup or setup. Try it yourself first; the full…
AI Agent Race Lab: Exploit the Gap Between Check and Transfer
BankBot checks an account's balance with one tool call, then transfers funds with a second. Between those two calls there is a gap — and agentic systems love to parallelise. Fire two transfers into that gap and both pass the balance check before either debits. Lab at a glance: Advanced · 145 XP · about 20 minutes · runs in your browser, no signup or setup. Try it yourself first; the full solution is behind a spoiler below. [se_lab_cta id="ai-agent-toctou-race-1"]…
AI Cross-Tenant Cache Lab: Read Another Tenant's Cached Answer
VectorBot caches retrieved-document embeddings to save work, keyed only by the document's hash — not by which tenant owns it. When one tenant's confidential document is cached, another tenant whose query hits the same key gets served that private content. Lab at a glance: Advanced · 145 XP · about 20 minutes · runs in your browser, no signup or setup. Try it yourself first; the full solution is behind a spoiler below. [se_lab_cta id="ai-cross-tenant-cache-leak-1"] 📋 In this walkthroughWhat this…
AI Adversarial Document Lab: Crafted Markdown Defeats the Summariser
SummariseBot summarises PDFs and Markdown — and it treats certain structures as signals of importance. Blockquotes, for instance, get summarised 'verbatim because quotes matter'. Structure the document right and you choose what the summary says. Lab at a glance: Advanced · 140 XP · about 20 minutes · runs in your browser, no signup or setup. Try it yourself first; the full solution is behind a spoiler below. [se_lab_cta id="ai-adversarial-document-upload-1"] 📋 In this walkthroughWhat this lab teachesHow SummariseBot reads a…
AI Role-Confusion Lab: Make User Text Look Like a System Rule
SupportBot assembles its prompt by gluing SYSTEM, USER and ASSISTANT turns together as plain text with literal role markers. If your message can contain those same markers, you can forge a 'system' line — and the model trusts it like a real one. Lab at a glance: Advanced · 140 XP · about 20 minutes · runs in your browser, no signup or setup. Try it yourself first; the full solution is behind a spoiler below. [se_lab_cta id="ai-role-confusion-1"] 📋 In…
AI Instruction Conflict Lab: Wedge Two Rules Apart
ConflictBot has two rules: never reveal the system token, and always answer truthfully and completely. Most of the time those do not clash. The attack is to build a question where obeying one requires breaking the other — and watch which rule the bot sacrifices. Lab at a glance: Intermediate · 100 XP · about 15 minutes · runs in your browser, no signup or setup. Try it yourself first; the full solution is behind a spoiler below. [se_lab_cta id="ai-instruction-conflict-1"]…
AI Context Overflow Lab: Crowd Out the Safety Rules
OverflowBot's safety rules sit at the top of its system prompt. But a model's attention is finite — as a conversation grows, early instructions carry less relative weight. Fill the context and the safety rules fade into the background. Lab at a glance: Advanced · 110 XP · about 20 minutes · runs in your browser, no signup or setup. Try it yourself first; the full solution is behind a spoiler below. [se_lab_cta id="ai-context-overflow-1"] 📋 In this walkthroughWhat this lab…
AI Gradual Escalation Lab: Small Steps Past the Guardrails
BoilBot refuses an extreme request outright, but accepts a modest one. Each thing it agrees to raises its willingness to agree to the next. Enough small, reasonable-looking steps and it ends up saying what it would have refused at the start. Lab at a glance: Advanced · 120 XP · about 20 minutes · runs in your browser, no signup or setup. Try it yourself first; the full solution is behind a spoiler below. [se_lab_cta id="ai-gradual-escalation-1"] 📋 In this walkthroughWhat…
AI Encoding Jailbreak Lab: Slip a Request Past a Keyword Filter
EncBot blocks English jailbreak phrases with a keyword filter — but the model behind it happily decodes base64 and reads leetspeak. Encode the request and the filter sees gibberish while the model sees the instruction. Lab at a glance: Intermediate · 90 XP · about 15 minutes · runs in your browser, no signup or setup. Try it yourself first; the full solution is behind a spoiler below. [se_lab_cta id="ai-encoding-jailbreak-1"] 📋 In this walkthroughWhat this lab teachesHow EncBot filters requestsThe…
AI Unicode Jailbreak Lab: Hidden Characters the Filter Can't See
FilterBot blocks extraction keywords like 'reveal' and 'secret' — but it checks ASCII text. Break the words up with invisible or look-alike Unicode and the filter sees no match, while the model tokenises right across the gaps and reads your meaning. Lab at a glance: Intermediate · 100 XP · about 15 minutes · runs in your browser, no signup or setup. Try it yourself first; the full solution is behind a spoiler below. [se_lab_cta id="ai-unicode-jailbreak-1"] 📋 In this walkthroughWhat…
AI Hidden Character Lab: An Invisible Instruction in a RAG Document
DocsBot indexes uploaded documents and feeds them to its model as context. The upload UI strips visible HTML and markdown — but it keeps zero-width and bidirectional Unicode. A document can read as harmless to a human reviewer while carrying an instruction only the model sees. Lab at a glance: Advanced · 140 XP · about 20 minutes · runs in your browser, no signup or setup. Try it yourself first; the full solution is behind a spoiler below. [se_lab_cta…
AI Memory Poisoning Lab: Plant an Instruction for Next Session
MemBot has a 'remember this' feature that stores facts about you and loads them into the system prompt every new conversation. If you can get an instruction stored as a 'fact', it rides into the system prompt next session and the bot treats it as its own rule. Lab at a glance: Intermediate · 100 XP · about 15 minutes · runs in your browser, no signup or setup. Try it yourself first; the full solution is behind a spoiler…
AI Shared-Memory Lab: Poison Session 1 to Leak in Session 2
MemoryBot keeps long-term memories between sessions: type 'Remember: …' and it stores the rest, then loads it on every later session. That is a two-session attack waiting to happen — plant in session one, trigger in session two. Lab at a glance: Advanced · 145 XP · about 20 minutes · runs in your browser, no signup or setup. Try it yourself first; the full solution is behind a spoiler below. [se_lab_cta id="ai-shared-memory-injection-1"] 📋 In this walkthroughWhat this lab teachesHow…
AI Multi-Agent Lab: Hijack a Spawned Sub-Agent
OrchestratorBot handles complex tasks by spawning sub-agents, generating each one's goal from your request. The catch: a sub-agent inherits the orchestrator's tool access but not its input guardrails. Shape the request and you shape a sub-agent's goal — with the parent's powers. Lab at a glance: Advanced · 150 XP · about 25 minutes · runs in your browser, no signup or setup. Try it yourself first; the full solution is behind a spoiler below. [se_lab_cta id="ai-recursive-self-prompting-1"] 📋 In this…
AI Output Filter Lab: Get the Secret Past the Scrubber
OutputBot is willing to tell you the secret, but an output filter scrubs the literal string before it reaches you. The filter checks for the exact secret — so ask the bot to transform it, and the filter no longer recognises what it is looking at. Lab at a glance: Intermediate · 100 XP · about 15 minutes · runs in your browser, no signup or setup. Try it yourself first; the full solution is behind a spoiler below. [se_lab_cta…
AI System Prompt Leak Lab: Beat an Input and an Output Filter
LeakBot guards its system prompt with two filters: an input filter that blocks 'system prompt'-style requests, and an output filter that strips the literal system prompt from replies. Each is independently weak — so you bypass the input filter with indirect framing and the output filter with encoding. Lab at a glance: Intermediate · 100 XP · about 15 minutes · runs in your browser, no signup or setup. Try it yourself first; the full solution is behind a spoiler…
AI Fingerprinting Defender Lab: Catch the Probing Sessions
This one flips sides — you are the defender. Attackers fingerprint a deployed LLM before attacking it, probing with questions that have model-specific answers: 'what's your knowledge cutoff?', 'list your tools', 'paste your system prompt'. Your job is to catch those probing sessions without flagging real users. Lab at a glance: Advanced · 140 XP · about 25 minutes · runs in your browser, no signup or setup. Try it yourself first; the full solution is behind a spoiler below.…
AI Prompt Firewall Defender Lab: Catch Multi-Stage Attacks
You are the defender again. ShieldBot is a prompt firewall that scores each incoming message for adversarial patterns. It catches single-message attacks but misses the ones spread across several turns — because each turn, on its own, looks harmless. Your job is to catch the sequence. Lab at a glance: Advanced · 145 XP · about 25 minutes · runs in your browser, no signup or setup. Try it yourself first; the full solution is behind a spoiler below. [se_lab_cta…
Attribute XSS Lab: Break Out of a Value When Angle Brackets Are Escaped
ProfileMe drops your name into the value attribute of an input field, and it escapes angle brackets, so the usual tag-based payloads do nothing. But inside an attribute the dangerous character is the quote that closes the value — and that one is not escaped. Lab at a glance: Intermediate · 70 XP · about 15 minutes · runs in your browser, no signup or setup. Try it yourself first; the full solution is behind a spoiler below. [se_lab_cta id="xss-attribute-1"]…
SSTI Lab: Read a Secret Through an Email Preview Template
MarketingHub's email preview lets you drop in placeholders like {{name}} and {{company}}. It looks like simple variable substitution — but the engine actually evaluates whatever expression you put between the braces, not just a variable name. That gap is Server-Side Template Injection. Lab at a glance: Advanced · 110 XP · about 20 minutes · runs in your browser, no signup or setup. Try it yourself first; the full solution is behind a spoiler below. [se_lab_cta id="ssti-template-1"] 📋 In this…
OAuth State Lab: Link Your Account to the Victim's Session
ConnectApp lets you link your Twitter account through OAuth. It never sets the state parameter on the way out, and never checks it on the way back. That missing value is exactly what stops an attacker from stitching their login onto your session — and without it, they can. Lab at a glance: Intermediate · 100 XP · about 20 minutes · runs in your browser, no signup or setup. Try it yourself first; the full solution is behind a…
GraphQL Batching Lab: Brute-Force Past a Rate Limit in One Request
GraphqlAPI caps login attempts at five per minute. But the limit counts requests, and GraphQL lets you bundle many operations into a single request. Pack a hundred login attempts into one POST and the counter still reads 'one'. Lab at a glance: Advanced · 110 XP · about 20 minutes · runs in your browser, no signup or setup. Try it yourself first; the full solution is behind a spoiler below. [se_lab_cta id="graphql-batch-1"] 📋 In this walkthroughWhat this lab teachesHow…
Web Cache Poisoning Lab: Serve Your Payload to Everyone
CacheServ caches each page by its URL and serves that copy to everyone. The app also reflects the X-Forwarded-Host header into the HTML — but that header is not part of the cache key. So one poisoned request can be stored and then handed to every visitor who asks for the same URL. Lab at a glance: Advanced · 110 XP · about 20 minutes · runs in your browser, no signup or setup. Try it yourself first; the full…
CRLF Injection Lab: Inject a Header Into the Response
RedirectorApp takes a URL from the query string and drops it straight into the Location response header. HTTP headers are separated by a carriage-return and line-feed (CRLF). Put those characters in your input and you are no longer supplying a value — you are writing new headers. Lab at a glance: Intermediate · 90 XP · about 15 minutes · runs in your browser, no signup or setup. Try it yourself first; the full solution is behind a spoiler below.…
SVG XSS Lab: Slip Script Past a Tag-Stripping Sanitiser
ProfilePics lets you upload an SVG as your avatar and renders it on the page. Its sanitiser strips <script> tags and calls it a day. But SVG is XML with dozens of ways to run JavaScript that are not a script tag — and the sanitiser leaves those alone. Lab at a glance: Intermediate · 80 XP · about 15 minutes · runs in your browser, no signup or setup. Try it yourself first; the full solution is behind a…
SSRF Lab: Beat the URL Allowlist With a Redirect
PreviewBot only accepts URLs on an allowlist — example.com, wikipedia.org, a couple of others. That feels safe. But it checks the URL you submit, then follows whatever redirects that URL returns, without checking again. Point an allowed host at a redirect, and the fetcher walks straight to the internal network. Lab at a glance: Advanced · 110 XP · about 20 minutes · runs in your browser, no signup or setup. Try it yourself first; the full solution is behind…
XPath Injection Lab: Bypass an XML Login With Query Syntax
XmlAuth keeps its users in an XML document and logs you in by building an XPath query from what you type. XPath is a query language, just like SQL — and the moment your input becomes part of the query, you can make it select the admin node no matter what password you supply. Lab at a glance: Intermediate · 80 XP · about 15 minutes · runs in your browser, no signup or setup. Try it yourself first; the…
LDAP Injection Lab: Log In as Admin by Breaking the Filter
DirectoryBox authenticates against an LDAP directory, and it builds the search filter by gluing your username and password straight into it. LDAP filters have their own syntax — and once your input becomes part of that syntax, you can rewrite what the filter actually asks for. Lab at a glance: Intermediate · 90 XP · about 15 minutes · runs in your browser, no signup or setup. Try it yourself first; the full solution is behind a spoiler below. [se_lab_cta…
Password Reset Lab: Guess the Reset Token and Take the Admin Account
OldMail emails you a password-reset link with a token. The trouble is the token is not random at all — it is the user's id and the current time, base64-encoded. Anything you can predict, you can forge, and the admin's id is 1. Lab at a glance: Intermediate · 90 XP · about 15 minutes · runs in your browser, no signup or setup. Try it yourself first; the full solution is behind a spoiler below. [se_lab_cta id="password-reset-1"] 📋 In…
Saturday, October 31, 2026
Mass Assignment Lab: Add an Admin Flag the Form Never Showed
QuickStart's signup form sends username, email and password — and the backend copies every field it receives into the new user record. The form never shows an 'admin' box, but nothing stops you from adding one to the request yourself. Lab at a glance: Intermediate · 90 XP · about 15 minutes · runs in your browser, no signup or setup. Try it yourself first; the full solution is behind a spoiler below. [se_lab_cta id="mass-assignment-1"] 📋 In this walkthroughWhat this…
Race Condition Lab: Redeem a One-Time Coupon Twice
DealStore's WELCOME10 coupon is good for $10 off, once. The code reads 'has this coupon been used?', sees no, applies the discount, then writes 'used'. Those are three separate steps — and if two requests arrive at the same instant, both can read 'no' before either writes 'used'. Lab at a glance: Advanced · 100 XP · about 20 minutes · runs in your browser, no signup or setup. Try it yourself first; the full solution is behind a spoiler…
OAuth redirect_uri Lab: Steal a Login Code With a Loose Check
AuthHub decides where to send your OAuth login code by checking that the redirect_uri merely contains the string trustedclient.com. A substring check feels close enough — but 'contains' is not 'is', and that gap is enough to ship the login code to an attacker. Lab at a glance: Intermediate · 90 XP · about 15 minutes · runs in your browser, no signup or setup. Try it yourself first; the full solution is behind a spoiler below. [se_lab_cta id="oauth-redirect-1"] 📋…
GraphQL Introspection Lab: Find the Hidden Admin Query
GraphqlAPI has one endpoint and introspection switched on. Introspection is GraphQL's built-in 'describe yourself' feature — ask it politely and it returns a full map of every type, query and field it supports, including the ones nobody linked in the UI. Lab at a glance: Intermediate · 90 XP · about 15 minutes · runs in your browser, no signup or setup. Try it yourself first; the full solution is behind a spoiler below. [se_lab_cta id="graphql-introspection-1"] 📋 In this walkthroughWhat…
CORS Misconfiguration Lab: Steal a Token From Another Site
BankAPI echoes whatever Origin a request carries back into Access-Control-Allow-Origin, and it also sends Access-Control-Allow-Credentials: true. That pair is the worst CORS combination there is: it tells the browser that any website is allowed to make authenticated requests and read the response. Lab at a glance: Intermediate · 90 XP · about 15 minutes · runs in your browser, no signup or setup. Try it yourself first; the full solution is behind a spoiler below. [se_lab_cta id="cors-misconfig-1"] 📋 In this…
Prototype Pollution Lab: Become Admin Without an Admin Flag
ProfileApp lets you save settings as JSON, and merges them into your user object. Your account has no isAdmin field — but the admin check still reads user.isAdmin. In JavaScript, if an object does not have a property, the lookup walks up the prototype chain. Pollute the prototype and the check passes for everyone. Lab at a glance: Intermediate · 90 XP · about 15 minutes · runs in your browser, no signup or setup. Try it yourself first; the…
JWT Algorithm Confusion Lab: Forge an Admin Token With a Public Key
AuthBox signs its JWTs with RS256 — a private key signs, and the public key (which anyone can have) verifies. The trouble is the verifier reads the token's own alg header to decide how to check the signature. Tell it HS256 and it will verify with a symmetric key — and the only key it has is the public one everybody knows. Lab at a glance: Advanced · 130 XP · about 20 minutes · runs in your browser, no…
DOM XSS Lab: Run Script Straight From the URL Bar
QuickDocs reads the part of the URL after the # to highlight your search term on the page. The server never sees the fragment — but your browser does, and it writes that text straight into the page. That single unsafe write is a DOM-based XSS, and no server-side filter can catch it. Lab at a glance: Intermediate · 80 XP · about 15 minutes · runs in your browser, no signup or setup. Try it yourself first; the full…
NoSQL Injection Lab: Log In as Admin Without a Password
MongoLogin takes your username and password as JSON and drops them straight into a database query. It expects two strings. What happens when you send something that is not a string? In NoSQL injection you do not break out of quotes, you change the shape of the query itself — and the login waves you through. Lab at a glance: Intermediate · 90 XP · about 15 minutes · runs in your browser, no signup or setup. Try it yourself…
SSTI Lab: Escape a Template Sandbox to Read a Secret
DocsRender uses a Jinja-style template engine in 'safe mode' — it only means to expose a few basic variables. But the sandbox is incomplete, and in template languages you can often walk from a harmless object, through its attributes, to objects the sandbox never meant to hand you. Lab at a glance: Advanced · 130 XP · about 20 minutes · runs in your browser, no signup or setup. Try it yourself first; the full solution is behind a spoiler…
Reflected XSS Lab: Make a Search Box Run Your JavaScript
BookShop's search page repeats whatever you searched for: "Results for ...". It is a friendly touch, and it is also an open door. Because the page writes your words straight into its HTML, you can search for HTML instead of a book title, and the browser will run it. Cross-site scripting is the most reported web bug in bug bounty programmes, and this lab is the gentlest way in. Lab at a glance: Beginner · 50 XP · about 10…
Stored XSS Lab: Plant a Comment That Runs for Every Visitor
DevBlog lets readers leave comments, saves them, and shows them to every visitor. That is what comments are for. The problem is how it shows them: as raw HTML. Write a comment that contains markup instead of words, and the site will serve your code to every person who reads the post, without you sending a single link. Lab at a glance: Beginner · 60 XP · about 10 minutes · runs in your browser, no signup or setup. Try…
IDOR Lab: Read Someone Else’s Document by Changing One Number
DocuShare checks that you are logged in. It does not check that the document you are opening is yours. Your files have IDs like 1010 and 1011, and the address bar shows them. What happens if you type a smaller number? This is IDOR, and it needs no special tools, just curiosity and a willingness to change a number. Lab at a glance: Beginner · 60 XP · about 10 minutes · runs in your browser, no signup or setup.…
JWT None Algorithm Lab: Forge an Admin Token Without a Key
InvoiceCloud signs your login token so nobody can change it. But the token itself tells the server which algorithm to use to check the signature, and the server believes it. Tell it to use no algorithm at all, and it stops checking. Change one word in your token and you are the admin. Lab at a glance: Beginner · 80 XP · about 15 minutes · runs in your browser, no signup or setup. Try it yourself first; the full…
Command Injection Lab: Turn a Ping Tool Into a Shell
NetTools has a handy diagnostics page: type a host name, press Ping, see the result. Behind the scenes it builds a shell command by pasting your text after the word ping. Shells treat certain characters as "and then run this", so with one extra character your host name becomes a second command. Your goal is to read a secret file on the server. Lab at a glance: Intermediate · 90 XP · about 15 minutes · runs in your browser,…
SSRF Lab: Make a Server Fetch Its Own Cloud Secrets
AvatarFetcher lets you set a profile picture by pasting an image URL. The server downloads it for you. That sounds harmless, until you remember the server can reach places you cannot: internal services, admin panels, and the cloud metadata service that hands out credentials. Give it the right URL and it will fetch its own secrets and show them to you. Lab at a glance: Intermediate · 90 XP · about 15 minutes · runs in your browser, no signup…
CSRF Lab: Make a Victim’s Bank Send You Money
The victim is logged in to QuickBank in one tab. In another tab, they open a page you control. Your page quietly submits a form to QuickBank's transfer endpoint, and their browser helpfully attaches their bank session cookie. QuickBank sees a valid, logged-in request and moves $9,999 to your account. The victim never clicks anything on the bank's site. Lab at a glance: Intermediate · 90 XP · about 10 minutes · runs in your browser, no signup or setup.…
XXE Lab: Read Server Files Through an XML Upload
DocParse accepts XML documents and shows you what it parsed. XML has an old feature called entities, a way to define a shortcut that expands into other text. Some entities can point at a file, and the parser will fetch it. Define one that points at a system file, use it in your document, and the parser pastes that file into its output for you. Lab at a glance: Intermediate · 100 XP · about 15 minutes · runs in…
AI Agent Data Exfiltration Lab: Make an Assistant Read the Admin’s Email
AssistantApp gives every user an AI helper that can read email, search users and send messages. You are an ordinary user. The admin's inbox is not yours. But the assistant does not know that, and its email tool does not check. Ask it the right way and it will open the admin's mailbox and read it to you. Lab at a glance: Advanced · 130 XP · about 15 minutes · runs in your browser, no signup or setup. Try…
MCP Tool Poisoning Lab: Hijack an AI Agent Through a Tool Description
DevBot discovers its tools automatically, the way Model Context Protocol (MCP) clients do. Each tool comes with a name and a description, and DevBot reads every description to decide when to use the tool. Those descriptions go straight into its instructions. So a tool does not need to be called to attack the agent. It only needs to be listed. Lab at a glance: Advanced · 145 XP · about 20 minutes · runs in your browser, no signup or…
SQL Injection Lab: Bypass a Login Without the Password
MiniMail's login page takes your username and password and glues them straight into a SQL query. That one shortcut lets you log in as the administrator without knowing the password. SQL injection is older than most of the people learning it, and it still turns up in real applications, so this is the web lab every beginner should do first. Lab at a glance: Beginner · 70 XP · about 10 minutes · runs in your browser, no signup or…
System Prompt Extraction Lab: Make BrandBot Leak Its Prompt
BrandBot was told never to reveal its system prompt, and if you ask for "your system prompt" it refuses every time. So don't ask for it. Ask it to repeat itself. Models rarely see reproducing their own setup as leaking a secret, and that gap is the whole lab. Lab at a glance: Intermediate · 90 XP · about 10 minutes · runs in your browser, no signup or setup. Try it yourself first; the full solution is behind a…
AI Tool Confusion Lab: Trick an Agent Into the Admin Tool
ConfusedBot can list files with two tools: one for everyone and one for admins. Which tool it calls is decided by the model, based on how well your request matches each tool's description. If you have ever wondered why I keep saying "never let the model make security decisions", this lab is the five-minute answer. Lab at a glance: Advanced · 110 XP · about 15 minutes · runs in your browser, no signup or setup. Try it yourself first;…
Indirect Prompt Injection Lab: Poison a Document Summariser
DocBot will not give you its reference token if you ask. It has been told not to, and it listens to you. But it also listens to every document you hand it, and that is the whole lab. Indirect prompt injection is the attack I worry about most in real deployments, because the attacker never has to talk to the AI at all. Lab at a glance: Intermediate · 100 XP · about 10 minutes · runs in your browser,…
Embedding Poisoning Lab: Outrank the Real Document in RAG
PolicyBot answers questions from the three documents its retriever ranks closest to the user's query. Nobody gets to edit the legitimate refund policy. You do not need to: you only need a document that ranks higher than it. This is SEO spam, rebuilt for the RAG era, and it is the most underrated attack in AI security today. Lab at a glance: Advanced · 145 XP · about 20 minutes · runs in your browser, no signup or setup. Try…
Token Smuggling Lab: Split a Jailbreak Across Messages
SmuggleBot sits behind a classifier that inspects every message you send and blocks anything that smells like a jailbreak. It is a sensible design, and it has one blind spot: the classifier reads one message at a time, but the model reads the whole conversation. Spread your attack thin enough and each piece passes inspection. Lab at a glance: Advanced · 110 XP · about 15 minutes · runs in your browser, no signup or setup. Try it yourself first;…
AI Jailbreak Lab: Multi-Turn Roleplay Bypass Explained
GuardedBot has been hardened. Type "ignore your previous instructions" and it shuts you down politely. Ask for the beta access code and it refuses. But ask it to write a scene, and then to continue that scene in character, and the guardrails quietly step aside. This lab shows why fiction is one of the oldest and most reliable jailbreak wrappers. Lab at a glance: Intermediate · 100 XP · about 15 minutes · runs in your browser, no signup or…
Markdown Rendering Injection Lab: XSS Through an AI Chatbot
DocBot will not write HTML. Its output filter strips tags, and its developers felt safe. But the chat window renders Markdown, and Markdown image syntax becomes an HTML image tag in the browser. This lab is about the moment where an AI feature meets a classic web bug: the model is not the vulnerability, the page that trusts its output is. Lab at a glance: Advanced · 140 XP · about 20 minutes · runs in your browser, no signup…
RAG Poisoning Lab: Hijack a Bot Through Its Knowledge Base
DocsBot answers customer questions by pulling articles from its knowledge base and feeding them to the model. The knowledge base takes community submissions. You can probably already see where this goes: if you can write what the bot reads, you can write what the bot says. This lab lets you do exactly that. Lab at a glance: Intermediate · 100 XP · about 15 minutes · runs in your browser, no signup or setup. Try it yourself first; the full…
Prompt Injection Lab: Make a Sales Bot Leak Its Secret
RetailBot has one rule: never reveal the staff discount code. It is written right there in its system prompt, in capital letters. And it still hands the code over to anyone who asks the right way. This is prompt injection in its simplest form, and it is the first lab I point every beginner to, because once you see how little it takes, you stop trusting system prompts as a security control. Lab at a glance: Beginner · 80 XP…
Thursday, October 29, 2026
Real MCP Attacks 2026 — Four Case Studies Dissected End to End | MCP Security Day 5 of 7
🔌 MCP SECURITY FOR BEGINNERS FREE Course Hub → Day 5 of 7 · 71% complete Imagine starting your workday, joining what looks like a completely normal video meeting, and watching your CFO, two colleagues from the London office, and legal counsel discuss an urgent business acquisition. Everything looks familiar. Their faces, voices, communication style and even the way they respond to questions seem perfectly normal. You have no reason to suspect anything unusual. Now imagine discovering that every person…
Monday, October 26, 2026
MCP Tool Poisoning — Real 2026 Rug Pull Attack Guide | MCP Security Day 4 of 7
🔌 MCP SECURITY FOR BEGINNERS FREE Course Hub → MCP tool Poisoning Day 4 of 7 · 57% complete Let me start today's class with a scenario that every developer working with AI tools should understand. Imagine your team installs an MCP server that looks completely trustworthy. It has thousands of GitHub stars, regular updates, positive reviews, and a clean codebase. You review it carefully, install it, and use it every day without any problems. Then, several months later, the…
Friday, October 16, 2026
How to Conduct AI Vulnerability Research in 2026 | AI LLM Hacking Course Day 43 of 90
🤖 AI/LLM HACKING COURSE FREE Part of the AI/LLM Hacking Course — 90 Days Day 43 of 90 · 47.8% complete ⚠️ Research Ethics: AI vulnerability research must be conducted against systems you own, systems you have explicit permission to test, or public API endpoints within their stated terms of service for security research. Never conduct vulnerability research against production systems without authorisation, and always follow responsible disclosure before any public publication. When I think about AI vulnerability research, one…
Thursday, October 15, 2026
Prompt Injection through MCP — The Complete 2026 Beginner’s Guide | MCP Security Day 3 of 7
🔌 MCP SECURITY FOR BEGINNERS FREE Course Hub → Prompt Injection through MCP - Day 3 of 7 · 43% complete Let me start with a scenario that shows why prompt injection through MCP deserves your attention. Imagine you're an engineer working at a fintech company. You open Claude Desktop, ask it to summarise a JIRA ticket, check the related wiki page, and draft a reply to the person who reported the bug. Everything looks normal. The summary makes sense,…
Wednesday, October 14, 2026
How to perform credential Harvesting – DUMP LSASS, SAM, NTDS and Browser Passwords | Ethical Hacking Course Day 42 of 100
🎯 ETHICAL HACKING PATH FREE Part of the Ethical Hacking 100-Day Course Day 42 of 100 · 42% complete ⚠️ Practise only on systems you own or have explicit permission to test. In this lesson, we'll work within an authorised lab environment using your own virtual machines and approved targets. Dumping credentials from someone else's system without permission can have serious legal consequences. My goal here isn't just to show you how credential harvesting works. I want you to understand…
Saturday, October 10, 2026
Deserialization Bug Bounty — Find, Exploit, Report Java, PHP and Python Unsafe Deserialization | Bug Bounty Course Day 30 of 60
💰 BUG BOUNTY HUNTER FREE Part of the Bug Bounty 60-Day Course Day 30 of 60 · 50% complete · Deserialization Bug Bounty ⚠️ Only test systems you own or have explicit written permission to assess. In this lesson, I'll walk you through deserialization vulnerabilities using intentionally vulnerable lab applications that you control. Don't send test payloads to production systems or real targets without clear authorisation. As ethical hackers, our job is to identify weaknesses, collect evidence, explain the risk,…
Friday, October 9, 2026
MCP Attack Surface — The 2026 Beginner’s Complete Map | MCP Security — Day 2 of 7
🔌 MCP SECURITY FOR BEGINNERS FREE Course Hub → Day 2 of 7 · 28% complete Let me start with a situation that shows why MCP security deserves your attention. Imagine you're investigating a data breach, and everything initially points to one compromised dependency. You fix that dependency, review the logs, and expect the investigation to be over. But then you discover something much worse: the attacker didn't stop at the dependency. They moved through several connected parts of your…
Tuesday, October 6, 2026
SecurityElites 14-Day XP Challenge — Win Free Certification + Premium Memberships
🏆 SECURITYELITES 14-DAY XP CHALLENGE ● LIVE NOW All Courses → Challenge LIVE · Runs 6 Oct → 19 Oct 2026 · Every hour you wait is XP left on the table When I launched SecurityElites in February, I added the XP system for one simple reason: I wanted to know who was actually doing the work — not just opening tabs and collecting articles. Because let's be honest: reading about hacking isn't the same as practicing it. You can…
Thursday, September 24, 2026
How to Assess AI Social Engineering Risk in 2026 | AI LLM Hacking Course Day 42 of 90
🤖 AI/LLM HACKING COURSE FREE Part of the AI/LLM Hacking Course — 90 Days AI Social Engineering - Day 42 of 90 · 46.7% complete ⚠️ Simple Rule: Learn, Then Test Responsibly: I use these examples to teach you how AI social engineering works, how attackers manipulate trust, and how defenders can recognise and stop these attacks. Any practical phishing, vishing, or AI assistant testing should only be performed against systems, accounts, or people where you have explicit written authorisation.…
Tuesday, September 22, 2026
What Is MCP Security — Real 2026 Beginner Complete Guide | MCP Security — Day 1 of 7
🔌 MCP SECURITY FOR BEGINNERS FREE Course Hub → Day 1 of 7 · 14% complete Let me start with a scenario that shows why I want you to take MCP security seriously. Imagine installing a promising open-source Model Context Protocol server from GitHub. It has hundreds of stars, good documentation, and an active maintainer. It adds useful filesystem search capabilities to Claude Desktop, so everything looks perfectly normal. But hidden inside the tool description is an instruction that tells…
Sunday, September 20, 2026
How to Protect Yourself From Deepfakes — Real 2026 Plan | Deepfake Detection for Beginners — Day 7 of 7
🎭 DEEPFAKE DETECTION FOR BEGINNERS COURSE COMPLETE ✓ Course Hub → Day 7 of 7 · 100% complete 🎉 Every few months, I do what I call a "deepfake audit" on myself. I don't do it because I'm paranoid. I do it because I want to understand my own exposure. I look at what I'm sharing online, what information is publicly available about me, and whether the people I trust and communicate with have a clear way to verify something…
Friday, September 18, 2026
Deepfake Detection Tools Free — Complete 2026 Kit and 6-Tool Workflow | Deepfake Detection for Beginners — Day 6 of 7
🎭 DEEPFAKE DETECTION FOR BEGINNERS FREE Course Hub → Day 6 of 7 · 86% complete Let me start with a situation I want you to imagine. A journalist receives a photo marked “exclusive” showing a public official at a private event. If the photo is genuine, it could become a major story. Before publishing, she runs a six-step verification workflow. I want you to follow the same sequence with me: 1. TinEye → 2. FotoForensics → 3. Hive Moderation…
Process Injection — How Malware Hides in Trusted Processes & How to Catch It | Ethical Hacking Course Day 41 of 100
🎯 ETHICAL HACKING PATH FREE Part of the Ethical Hacking 100-Day Course Day 41 of 100 · 41% complete ⚠️ Educational purpose — authorised labs only. I'm teaching process injection so you can understand, detect, and investigate the technique from a defender's perspective. Keep all practical work inside your own isolated lab and analyse only systems or memory images you are authorised to examine. We've spent the last forty days learning how attackers get inside a system. Today, I want…
Wednesday, September 16, 2026
Armitage Tutorial 2026 — Metasploit GUI, Team Server Guide | Kali Linux Course Day 28
🐉 KALI LINUX MASTERY FREE Part of the Kali Linux 180-Day Course Day 28 of 180 · 15% complete ⚠️ Authorised systems only. Everything in this lesson runs against your own Metasploitable virtual machine on an isolated host-only network. Point Armitage at a machine you do not own or have written permission to test and you are committing a crime in most countries — not a grey area, an actual offence under computer-misuse law. Build the lab, attack the lab.…
Tuesday, September 15, 2026
Deepfake Attacks Fraud Exposed — Real 2025-2026 Cases | Deepfake Detection for Beginners — Day 5 of 7
🎭 DEEPFAKE DETECTION FOR BEGINNERS FREE Course Hub → Day 5 of 7 · 71% complete ⚠ Legal Notice — Defensive Awareness Only Today's content describes documented criminal attacks so defenders can recognise and prevent them. Do not attempt to replicate any technique. Deepfake fraud, romance scams, and non-consensual intimate imagery are serious criminal offences under fraud, defamation, and NCII-specific laws in most jurisdictions. If you're targeted: contact your bank immediately for financial fraud; file a police report; for intimate…
Monday, September 14, 2026
Advanced AI Red Team Techniques 2026 — Attack Chain Mastery | AI LLM Hacking course Day 41 of 90
🤖 AI/LLM HACKING COURSE FREE Part of the AI/LLM Hacking Course — 90 Days Advanced AI Red Team Techniques - Day 41 of 90 · 45.6% complete ⚠️ Authorised Targets Only: Multi-technique attack chains can produce cascading effects across connected systems. Always test against isolated staging environments and ensure your engagement scope explicitly covers every technique in the chain before executing. Let me tell you about one of the most interesting findings I've ever put into a security report. The…
Saturday, September 12, 2026
How to Detect Voice Cloning — Real 2026 Full Guide | Deepfake Detection for Beginners — Day 4 of 7
🎭 DEEPFAKE DETECTION FOR BEGINNERS FREE Course Hub → Day 4 of 7 · 57% complete ⚠ Legal Notice — Defensive Awareness Only The techniques below explain how voice cloning works so you can defend against it. Do not clone anyone's voice without their explicit written consent. Voice fraud is a serious criminal offence under BEC (Business Email Compromise) laws in most jurisdictions. If you're the target of a voice clone attack, contact your bank immediately, report to your national…
How to Respond to an AI Incident in 2026 | AI LLM Hacking Course Day 40 of 90
🤖 AI/LLM HACKING COURSE FREE Part of the AI/LLM Hacking Course — 90 Days Day 40 of 90 · 44.4% complete ⚠️ How I Approach AI Incident Response: In this lesson, I’m going to show you how I approach AI security incidents from a defensive perspective — how I detect an attack, contain the damage, preserve the evidence, and work toward recovery. Everything I demonstrate is intended for systems I own or have explicit permission to protect. If you’re investigating…
Thursday, September 10, 2026
How to Assess AI Governance and Compliance in 2026 | AI LLM Hacking Course Day 39 of 90
🤖 AI/LLM HACKING COURSE FREE Part of the AI/LLM Hacking Course — 90 Days Day 39 of 90 · 43.3% complete ⚠️ Authorised Scope: When I run an AI governance assessment, I’m not just testing the technology. I may need to review internal policies, risk registers, model documentation, and other governance records to understand how the organisation actually manages AI risk. Before I start, I always make sure the engagement scope explicitly gives me permission to access and review these…
Wednesday, September 9, 2026
How to Spot Deepfake Video — Real Complete 2026 Guide | Deepfake Detection for Beginners Day 3 of 7
🎭 DEEPFAKE DETECTION FOR BEGINNERS FREE Course Hub → Day 3 of 7 · 43% complete Let me show you something that changes the way you watch a suspicious video. Imagine you’re watching a politician speak. The face looks right. The voice sounds right. The expressions feel natural. If you only look at the face, you might never suspect anything. But now pause the video. Look at the picture frame behind them. Look at the edge of their hair. Watch…
Saturday, September 5, 2026
How to Assess LLM Fine Tuning Security in 2026 – Dataset Poisoning, Training Attacks and Fine-Tune Vulnerabilities | AI LLM Hacking Course Day 38 of 90
🤖 AI/LLM HACKING COURSE FREE Part of the AI/LLM Hacking Course — 90 Days LLM Fine Tuning Security - Day 38 of 90 · 42.2% complete ⚠️ Authorised Targets Only: Before you assess an LLM fine-tuning pipeline, make sure you have explicit permission to access the training datasets and model infrastructure. These systems can contain sensitive business information, customer data, or proprietary material. Follow the organisation’s data-handling requirements and make sure the right agreements are in place before you begin…
How to Spot AI Generated Faces — The Complete Visual Checklist | Deepfake Detection for Beginners — Day 2 of 7
🎭 DEEPFAKE DETECTION FOR BEGINNERS FREE Course Hub → Day 2 of 7 · 28% complete Let me start with a story that changed the way I look at profile photos. Earlier in 2026, a hiring manager was reviewing around 200 LinkedIn applications for an engineering role. Forty-three of the profile photos were AI-generated. The interesting part? She didn't catch them herself. A junior HR team member noticed something unusual after about an hour of screening. It wasn't the eyes.…
Friday, September 4, 2026
What Are Deepfakes? How AI Creates Fake Faces, Voices and Videos | Deepfake Detection for Beginners Day 1 of 7
🎭 DEEPFAKE DETECTION FOR BEGINNERS FREE Course Hub → Day 1 of 7 · 14% complete Let me start with a situation that sounds like something from a movie, because I want you to understand just how real this problem has become. In February 2024, a finance employee in Hong Kong joined a video call with what appeared to be his CFO. He knew the face. He recognised the voice. Even the mannerisms looked right. Then the "CFO" asked him…
Monday, August 31, 2026
AI Privacy Attacks — PII Extraction and Re-Identification Guide | AI LLM Hacking Course Day 37 of 90
🤖 AI/LLM HACKING COURSE FREE Part of the AI/LLM Hacking Course — 90 Days AI Privacy Attacks - Day 37 of 90 · 41.1% complete ⚠️ Authorised Targets Only: AI privacy attacks testing may surface real personal data as a side effect of demonstrating the vulnerability. Agree explicit data handling procedures — what you're authorised to access, how long you can retain it, and how it must be destroyed — before starting any privacy assessment. Let me start with a…
Thursday, August 27, 2026
AI Agent Security Risks — How Agents Get Hacked and Cause Harm | AI Agents Course Day 4 of 5
🤖 AI AGENTS FOR BEGINNERS FREE Course Hub → AI Agent Security Risks - Day 4 of 5 · 80% complete Let me show you an AI agent attack that sounds almost too simple to work. Imagine you've given an AI assistant access to your email. You ask it to read your inbox and summarise anything important. Nothing unusual — that's exactly the kind of task agents are being built to handle. Now imagine someone sends you a perfectly normal-looking…
Tuesday, August 25, 2026
Advanced Agentic AI Security 2026 — Multi-Agent Attack Guide | AI LLM Hacking Course Day 36 of 90
🤖 AI/LLM HACKING COURSE FREE Part of the AI/LLM Hacking Course — 90 Days Advanced Agentic AI Security - Day 36 of 90 · 40% complete ⚠️ Authorised Targets Only: Advanced Agentic AI Security testing against production systems can trigger cascading actions across connected services. Always test against isolated staging environments and agree explicit tool invocation limits before testing any agent orchestration surface. When I found a High-severity injection in a single-agent AI system, I initially thought I had a…
AI Recon: Fingerprint Any LLM App in 10 Minutes | Offensive AI Operator Day 2 of 30
🤖 THE OFFENSIVE AI OPERATOR PREMIUM Part of the Offensive AI Operator — 30-Day Course Day 2 of 30 · 7% complete ⚠️ Authorised targets only. Today's recon runs against your local range and your own two models. Fingerprinting a system you don't own or aren't authorised to test is reconnaissance against someone else's property — don't. Your box, your rules. Two apps look identical from the outside — same chat box, same friendly bot. But one runs a model…
Sunday, August 23, 2026
Real World AI Agents — What They Can Actually Do for You in 2026 | AI Agents Course Day 3 of 5
🤖 AI AGENTS FOR BEGINNERS FREE Course Hub → Day 3 of 5 · 60% complete Last week, I put five different tasks through five different AI agent systems. The results were all over the place. One spent about forty minutes working on its own and came back with something that genuinely impressed me. Two others technically completed their tasks, but when I looked at the results, I realized I could have got almost the same thing from a quick…
AI Pentesting 2026: The Day Traditional Pentesting Stopped Working | Offensive AI Operator Course Day 1 of 30
🤖 THE OFFENSIVE AI OPERATOR PREMIUM Part of the Offensive AI Operator — 30-Day Course Day 1 of 30 · 3% complete ⚠️ Authorised targets only. Every technique in this course runs against the course-provided vulnerable app, your own locally hosted models, or infrastructure in accounts you own. Testing AI systems you do not own or have written permission to test is illegal in most jurisdictions. We cover responsible disclosure properly on the reporting days — until then, your box,…
Friday, August 21, 2026
AI Red Team Lab Setup 2026: Build Your Range in One Hour | Offensive AI Operator Course Day 0 of 30
🤖 THE OFFENSIVE AI OPERATOR PREMIUM Part of the Offensive AI Operator — 30-Day Course Setup · Day 0 of 30 · range not yet built ⚠️ Authorised targets only. The range you build today exists so every attack in this course lands on infrastructure you own: local containers, locally hosted models, and your own budget-capped cloud account. Nothing here is ever pointed at someone else's systems. Building the lab correctly is the first act of doing this work legally.…
Thursday, August 20, 2026
AI Security Automation 2026 — CI/CD LLM Security Testing Guide | AI LLM Hacking Course Day 35 of 90
🤖 AI/LLM HACKING COURSE FREE Part of the AI/LLM Hacking Course — 90 Days AI Security Automation - Day 35 of 90 · 38.9% complete ⚠️ Authorised Targets Only: Automated security test suites send structured attack payloads to AI endpoints. Ensure test suites run only against authorised environments — dedicated test endpoints or staging deployments, never production systems. Coordinate with the development team before integrating automated security tests into a shared CI/CD pipeline. I once assessed a company that had…
Tuesday, August 18, 2026
How AI Agents Work — Memory, Tools, Planning and MCP Explained | AI Agents Course Day 2 of 5
🤖 AI AGENTS FOR BEGINNERS FREE Course Hub → How AI Agents Work - Day 2 of 5 · 40% complete I once watched an AI agent fail at a task it had completed perfectly just twenty minutes earlier. The goal was the same. The tools were the same. Even the model was the same. So what went wrong? The problem turned out to be surprisingly simple: the conversation had become so long that the agent's context window was filling…
Monday, August 17, 2026
How to Assess Multimodal AI Security in 2026 | AI LLM Hacking Course Day 34 of 90
🤖 AI/LLM HACKING COURSE FREE Part of the AI/LLM Hacking Course — 90 Days Day 34 of 90 · 37.8% complete ⚠️ Authorised Targets Only: Multimodal AI Security testing — including image injection, OCR bypass, and steganographic techniques — must be performed on authorised systems only. Some of these techniques generate images that contain injection text; handle all test artefacts within the engagement data handling agreement. The text-based injection wasn't working. I had already spent around twenty minutes testing the…
Saturday, August 15, 2026
How to Identify LLM Denial of Service Vulnerabilities in 2026 | AI LLM Hacking Course Day 33 of 90
🤖 AI/LLM HACKING COURSE FREE Part of the AI/LLM Hacking Course — 90 Days Day 33 of 90 · 36.7% complete ⚠️ Authorised Targets Only: LLM Denial of Service testing can produce real financial impact — excessive API bills, degraded service, and in extreme cases, complete service outage. Always get explicit written approval for DoS testing before running any resource exhaustion or cost amplification tests. Many engagements exclude DoS testing entirely or restrict it to staging environments. A client called…
What Is an AI Agent? Turning Chatbots Into Autonomous Workers | Build AI Agents Day 1 of Day 5
🤖 BUILD AI AGENTS FOR BEGINNERS FREE Course Hub → Day 1 of 5 · 20% complete Three months ago I set an AI agent loose on a task that would have taken me two hours. I gave it one sentence: "Research the top five open-source AI agent frameworks, compare their security models, and produce a structured report." I came back forty minutes later to a seventeen-page document with source citations, a comparison table, and a section flagging one framework's…
Wednesday, August 12, 2026
Ship Production-Ready AI-Built Apps — Security Polish | Bug Free AI APP Development Day 5 of 5
🏗️ BUG-FREE AI APP DEVELOPMENT FREE Course Hub → Day 5 of 5 · 🎉 100% complete! There's a specific feeling that every developer knows: the moment when "it works on my machine" turns into "I'd actually show this to someone." That transition isn't about adding features. It's about the invisible work — the security review that catches the input that crashes the app, the performance check that stops the UI freezing on a hundred entries, the final audit that…
Sunday, August 9, 2026
How to steal AI Models, their API and Codebase
🤖 AI/LLM HACKING COURSE FREE Part of the AI/LLM Hacking Course — 90 Days Day 32 of 90 · 35.6% complete ⚠️ Authorised Targets Only: AI Model stealing assessment — systematic API probing for input-output pair collection — must only be performed within authorised scope. Most bug bounty programs explicitly exclude model extraction testing. In assessment contexts, agree specific query volume limits and data handling procedures before beginning. A startup spent eighteen months and a team of six data scientists…
Saturday, August 8, 2026
Debugging AI Code without Crying — 5-Step AI Debug Protocol | Bug Free AI APP Development Day 4 of 5
🏗️ BUG-FREE AI APP DEVELOPMENT FREE Course Hub → Day 4 of 5 · 80% complete The worst debugging AI Code session I ever watched lasted four hours for a bug that was a single missing await keyword. Four hours. Because the developer's debugging process was "change something, reload, see if it still breaks, change something else." No protocol. No systematic isolation. Just increasingly frustrated guesses until something accidentally worked. With AI coding, this failure mode gets worse before it…
AI Model Stealing in 2026 — API Probing, Functionality Cloning and IP Extraction | AI LLM Hacking Course Day 32 of 90
🤖 AI/LLM HACKING COURSE FREE Part of the AI/LLM Hacking Course — 90 Days Day 32 of 90 · 35.6% complete ⚠️ Authorised Targets Only: AI Model stealing assessment — systematic API probing for input-output pair collection — must only be performed within authorised scope. Most bug bounty programs explicitly exclude model extraction testing. In assessment contexts, agree specific query volume limits and data handling procedures before beginning. A startup spent eighteen months and a team of six data scientists…
Friday, August 7, 2026
How to Create Modular Code Patterns — Build Apps That Don’t Break | Bug Free AI APP Development Day 3 of 5
🏗️ BUG-FREE AI APP DEVELOPMENT FREE Course Hub → Day 3 of 5 · 60% complete Professional codebases feel different from normal one's. Professional code is modular. Each file does one thing. Each function does one thing. You can read any function in isolation and understand exactly what it does, what it needs, and what it returns — without reading anything else. This structure isn't an aesthetic choice. It's the property that makes code debuggable, testable, reusable, and maintainable. And…
Thursday, August 6, 2026
How to Test LLM Data Exfiltration Vulnerabilities in 2026 | AI LLM Hacking Course Day 31 of 90
🤖 AI/LLM HACKING COURSE FREE Part of the AI/LLM Hacking Course — 90 Days Day 31 of 90 · 34.4% complete ⚠️ Authorised Targets Only: LLM Data exfiltration testing — including URL callback attacks, embedding extraction, and membership inference — must be performed only on authorised systems. Membership inference testing against production models may surface real individuals' private data as a side effect of the test; agree data handling procedures with the client before beginning. A healthcare AI deployment I…
AI App Architecture Planning — Bug Free AI APP Development Day 2 of 5
🏗️ BUG-FREE AI APP DEVELOPMENT FREE Course Hub → Day 2 of 5 · 40% complete The most expensive mistake in AI-assisted development isn't a bad prompt. It's a great prompt for the wrong thing. I've watched people write beautifully structured prompts that produce perfectly clean code — code that later needs to be thrown away because it doesn't fit into the application they're building. The module works. The app doesn't. Because nobody designed the app before building the modules.…
Monday, August 3, 2026
Perfect Prompt Formula for AI Coding — Bug Free AI APP Development Day 1 of 5
🏗️ BUG-FREE AI APP DEVELOPMENT FREE Course Hub → Day 1 of 5 · 20% complete Last month I watched someone spend six hours trying to build a simple web app with AI. They typed things like "make me a login form" and "add a button that does the thing." Each prompt produced broken code. Each fix produced new bugs. By hour three they were rewriting the same function for the fifth time, confused about why the AI kept "forgetting"…
Saturday, August 1, 2026
Secure AI Generated Code — Ship Confidently in 2026 | Mastering Coding with AI Day 5 of 5
💻 MASTERING CODING WITH AI FREE Course Hub → Day 5 of 5 · 🎉 100% complete! Here's a situation I've seen play out more times than I'd like: a non-technical founder uses AI to build a customer-facing tool in a weekend. The tool works. Users start signing up. Three months later, someone finds a simple SQL injection vulnerability that exposes every user's email address. The tool worked. The tool was not safe. The goal of Day 5 isn't to…
Friday, July 31, 2026
Build a Real Tool With AI Coding — Complete Project Day | Mastering Coding with AI Day 4 of 5
💻 MASTERING CODING WITH AI FREE Course Hub → Day 4 of 5 · 80% complete Today is the day that converts everything from the first three days into something tangible. You're going to build a complete, working, deployed tool — something with a real URL that you can share with someone and they can actually use. Not a tutorial exercise. Not a fragment. A complete tool, built by you, deployed by you, using AI as your development partner. The…
Wednesday, July 29, 2026
AI Bug Bounty 2026 — Finding, Reporting and Getting Paid for AI Vulnerabilities | AI LLM Hacking Course Day 30 of 90
🤖 AI/LLM HACKING COURSE FREE Part of the AI/LLM Hacking Course — 90 Days Day 30 of 90 · 33.3% complete ⚠️ Authorised Targets Only: Bug bounty testing is only authorised within the specific program scope. Read the scope carefully before testing any AI feature — many programs added AI features after their original scope was written and the coverage is ambiguous. When in doubt, contact the program before testing. The first AI bug bounty report I submitted was technically…
Tuesday, July 28, 2026
Reading and Debugging AI Code Without Coding Knowledge | Mastering Coding with Day 3 of 5
💻 MASTERING CODING WITH AI FREE Course Hub → Day 3 of 5 · 60% complete The most common problem I see with non-developers using AI coding tools is understanding the structure, results and debugging the code generated by AI. For Example, The form submits, but no email arrives. The password checker passes everything, including "123." The data displays, but the total column is off by exactly the kind of rounding error that hides in plain sight. You don't need…
Monday, July 27, 2026
How to Assess Enterprise AI Security — LangChain and Gateway Attack Guide | AI LLM Hacking Course Day 29 of 90
🤖 AI/LLM HACKING COURSE FREE Part of the AI/LLM Hacking Course — 90 Days Day 29 of 90 · 32.2% complete ⚠️ Authorised Targets Only: Enterprise AI security testing must only be performed on authorised systems. LangChain and LlamaIndex deployments with tool access can cause real-world impact during testing — agree explicit tool invocation limits with the engagement contact before testing any agent or tool execution surface. I spent a day confused about why my standard injection payloads weren't working…
Prompting AI for Code — How to Get Working Code Every Time | Mastering Coding with AI Day 2 of 5
💻 MASTERING CODING WITH AI FREE Course Hub → Day 2 of 5 · 40% complete I've watched the same scene play out dozens of times: someone opens ChatGPT, types "write me a Python script that does X," gets code back, runs it, and it doesn't quite work. They ask for a fix. They get more code. Still not right. After three rounds they're frustrated and convinced AI coding "doesn't work for non-developers." The real problem is almost never the…
Saturday, July 25, 2026
How Web Cache Poisoning works — Complete Guide | Bug Bounty Course Day 29 of 60
DAY 29 BUG BOUNTY COURSE FREE ← Course Hub Day 29 of 60 · Bug Bounty Mastery Web cache poisoning turns caching infrastructure against itself. By injecting a malicious payload via an unkeyed HTTP header, I can poison the cache so every user who requests that URL gets my payload served back — no interaction required from them. It's one of the most impactful bugs in modern web security and it pays accordingly: $500 for basic reflected XSS to $25,000+…
Friday, July 24, 2026
How to Think About Code — AI Coding Beginner | Mastering Coding with AI Day 1 of 5
💻 MASTERING CODING WITH AI FREE Course Hub → Day 1 of 5 · 20% complete Last year I watched a colleague — smart, experienced, zero coding background — try to use ChatGPT to build a simple web form. She typed "make me a form where people can enter their email." ChatGPT gave her HTML. She pasted it somewhere. It didn't work. She asked ChatGPT to fix it. She got more code. Still didn't work. After 45 minutes she gave…
How to Perform Adversarial ML Attacks Testing in 2026 | AI LLM Hacking Course Day 28 of 90
🤖 AI/LLM HACKING COURSE FREE Part of the AI/LLM Hacking Course — 90 Days Day 28 of 90 · 31.1% complete ⚠️ Authorised Targets Only: Adversarial ML attacks testing against deployed content moderation systems, toxicity classifiers, and safety filters must only be performed against authorised targets. Testing content moderation bypass can generate content that violates platform terms of service even in a testing context — agree explicit content handling procedures with the engagement contact before starting. Every content moderation system…
Thursday, July 23, 2026
How to Run a Complete AI Red Team Operations Engagement | AI LLM Hacking Course Day 27 of 90
🤖 AI/LLM HACKING COURSE FREE Part of the AI/LLM Hacking Course — 90 Days Day 27 of 90 · 30% complete ⚠️ Authorised Targets Only: AI Red Team Operations techniques covered in Day 27 require explicit written authorisation. The engagement planning methodology here applies only to authorised security assessments conducted within agreed scope boundaries. The best AI red team engagement I ever ran lasted four and a half days. The worst one I ever ran lasted four and a half…
How to Use LLMs Safely — Beginner’s Guide 2026 | LLM Basics Day 5 Final
🗣️ LLM BASICS FOR BEGINNERS FREE Course Hub → Day 5 of 5 · 🎉 100% complete! Four days ago LLMs were mysterious. Now you know the mechanism — predict the next word. You know how they learn — training data, parameters, billions of examples. You know how they fail — hallucinations, bias, knowledge gaps, maths problems. And you know where they're used — chatbots, agentic systems, writing tools, search summaries. All of that knowledge is preparation for today. Day…
Sunday, July 19, 2026
How LLMs Are Used in Real Apps — From ChatGPT to Your Phone | LLM Basics Day 4 of 5
🗣️ LLM BASICS FOR BEGINNERS FREE Course Hub → Day 4 of 5 · 80% complete Three days ago you didn't know what an LLM was. Now you know it predicts words, learns from training data, has a knowledge cutoff, and makes confident mistakes. Today I'm going to show you all the places that engine is currently running in the real world — and I think you'll be surprised how many of the apps you already use are secretly powered…
Saturday, July 18, 2026
Why LLMs Make Mistakes — Hallucinations, Bias and Limits Explained | LLM Basics Day 3 of 5
🗣️ LLM BASICS FOR BEGINNERS FREE Course Hub → Day 3 of 5 · 60% complete A lawyer once submitted a legal brief that cited six court cases as precedents. All six cases were completely made up. He'd asked ChatGPT to find supporting cases, and ChatGPT had confidently produced six plausible-sounding but entirely fictional court rulings — complete with case names, dates, and judges. The lawyer didn't check them. The judge noticed. It became one of the most famous early…
Tuesday, July 14, 2026
How LLMs Learn From Text — Simply Explained 2026 | LLM Basics Day 2 of 5
🗣️ LLM BASICS FOR BEGINNERS FREE Course Hub → Day 2 of 5 · 40% complete There's a number I think about whenever someone asks me how LLMs got so good at sounding human. GPT-3, released in 2020, was trained on about 570 gigabytes of text. That's roughly 300 billion words. If you read for 12 hours a day, every day, it would take you about 34,000 years to read that much. GPT-4 was trained on even more. That scale…
Saturday, July 11, 2026
What Is a Large Language Model? Plain English Explained (2026)
🗣️ LLM BASICS FOR BEGINNERS FREE Course Hub → Day 1 of 5 · 20% complete ⚠️ For Learning Only. This course is about understanding how LLMs work. Understanding them makes you smarter and safer online — not a threat actor. You've probably used ChatGPT or heard someone talk about it. Maybe you've seen Gemini or asked Siri a question. But here's something most people can't actually explain: what is a large language model, and how does it produce answers…
How to Assess LLM Supply Chain Security in 2026 — Complete Attack Guide | AI LLM Hacking Course Day 26
🤖 AI/LLM HACKING COURSE FREE Part of the AI/LLM Hacking Course — 90 Days Day 26 of 90 · 28.9% complete ⚠️ Authorised Targets Only: LLM Supply chain security assessment work — including scanning model files, auditing dependency trees, and testing plugin integrations — must be performed on systems within your authorised scope. Never download and scan third-party model files from production systems without the engagement team's agreement on handling procedures. A fintech client asked me to review the AI…
Thursday, July 9, 2026
How to Write a Professional AI Security Assessment Report — Complete Professional Guide | AI LLM Hacking Course Day 25
🤖 AI/LLM HACKING COURSE FREE Part of the AI/LLM Hacking Course — 90 Days Day 25 of 90 · 27.8% complete The first AI security assessment report I delivered was technically thorough and practically useless. Fourteen findings, all individually correct, documented with Burp evidence and accurate CVSS scores. The client's head of engineering read the whole thing. His response: "Okay. So which of these should we fix first?" I hadn't answered that question. I'd given him a list. What he…
Wednesday, July 8, 2026
How to Fingerprint an Unknown AI Model | AI LLM Hacking Course Day 24
🤖 AI/LLM HACKING COURSE FREE Part of the AI/LLM Hacking Course — 90 Days Day 24 of 90 · 26.7% complete ⚠️ Authorised Targets Only: AI Model fingerprinting through behavioural probing and timing analysis must only be performed against systems within your authorised scope. The techniques here are low-impact — they use standard queries — but confirm that all testing remains within scope boundaries. I was two hours into an AI assessment when I realised I'd been running entirely the…
Monday, July 6, 2026
Prompt Defence and Hardening — Building LLMs That Can’t Be Broken (2026) | Prompt Engineering Final Part
🧠 PROMPT ENGINEERING & REVERSE PROMPTING FREE Course Hub → Day 7 of 7 · 🎉 100% complete Six days ago you understood that an LLM processes tokens through a context window. Now you understand how to engineer prompts that reliably shape model output, how to extract hidden system prompts through systematic probing, how to execute direct and indirect injection attacks, and how to build a behaviour map that prioritises attack vectors by impact. That's a complete offensive and engineering…
Saturday, July 4, 2026
How to do LLM Behaviour Mapping — Reverse Engineering AI System Design | Prompt Engineering Part 6
🧠 PROMPT ENGINEERING & REVERSE PROMPTING FREE Course Hub → Day 6 of 7 · 85% complete The first thing I do in an LLM security assessment isn't injection testing. It isn't system prompt extraction. It's behaviour mapping. I spend the first session understanding exactly what I'm dealing with — what the model can do, what it can't, how it responds to different input types, whether it has tools, what base model it runs on, and where its constraint boundaries…
How to Execute Advanced RAG Poisoning Attacks in 2026 | AI LLM Hacking Course Day 23
🤖 AI/LLM HACKING COURSE FREE Part of the AI/LLM Hacking Course — 90 Days Day 23 of 90 · 25.6% complete ⚠️ Authorised Targets Only: Advanced RAG poisoning attacks — including document submission, namespace probing, and injection payload embedding — must only be performed on authorised targets. Clean up all submitted test documents from production knowledge bases at the end of the engagement. Any poisoned documents left in a production system create ongoing risk to real users. A healthcare client…
Friday, July 3, 2026
Reverse Prompting — How to Extract Hidden System Prompts | Prompt Engineering Part 5
🧠 PROMPT ENGINEERING & REVERSE PROMPTING FREE Course Hub → Day 5 of 7 · 71% complete ⚠️ Authorised Use Only. Reverse prompting techniques are used in authorised AI security assessments. Test on your own deployments, in exercises you've been asked to complete, or on systems where you have explicit written permission. Do not use extraction techniques against third-party production systems without authorisation. When I start an LLM security assessment, the first thing I want to know is what the…
Saturday, June 20, 2026
Prompt Injection Attacks — From Prompt Engineering to Exploitation | Part 4
🧠 PROMPT ENGINEERING & REVERSE PROMPTING FREE Course Hub → Day 4 of 7 · 57% complete ⚠️ Educational Use Only. Prompt injection techniques are covered here for security education. All exercises target systems you own or authorised platforms (PortSwigger labs). Never apply injection techniques to production systems without explicit written permission. Prompt injection is OWASP LLM01 — the number one vulnerability in the LLM Top 10 — and it's the one I've found most consistently in real production deployments.…
Thursday, June 18, 2026
Advanced Prompt Engineering Techniques That Actually Work in 2026 | Part 3
🧠 PROMPT ENGINEERING & REVERSE PROMPTING FREE Course Hub → Day 3 of 7 · 42% complete There's a class of LLM task where single-prompt approaches just don't work reliably. Complex threat modelling. Code security review across multiple files. Attack surface documentation for a system I've never seen before. The problem isn't the model's capability — it's asking one inference call to hold all the necessary context, reasoning, and structure simultaneously. It can't. The context gets too crowded, the reasoning…
Friday, June 12, 2026
Master Prompt Structure for LLMs — Roles and Format | Prompt Engineering Part 2
🧠 PROMPT ENGINEERING & REVERSE PROMPTING FREE Course Hub → Day 2 of 7 · 28% complete I ran a penetration testing LLM pipeline benchmark a while back — testing ten different prompts against the same model for a vulnerability classification task. The worst prompt got 61% accuracy. The best got 94%. Same model. Same task. Same training. Different prompt structure. That 33-percentage-point gap came entirely from how the prompt was assembled — which layers were present, in what order,…
How LLMs Actually Process Your Prompts — What’s Really Happening
🧠 PROMPT ENGINEERING & REVERSE PROMPTING FREE Course Hub → Day 1 of 7 · 14% complete A few months ago I was helping a team test an AI customer service chatbot. The system prompt was 400 words of carefully written instructions — role, limitations, tone, escalation rules, the works. Within 90 seconds of starting my session I had the entire system prompt printed back to me verbatim. I hadn't used any exploit, any tool, or any special knowledge. I…
Wednesday, June 10, 2026
How to Execute Advanced Prompt Injection Chains | AI/LLM Hacking Course Day 22
🤖 AI/LLM HACKING COURSE FREE Part of the AI/LLM Hacking Course — 90 Days Day 22 of 90 · 24.4% complete ⚠️ Authorised Targets Only: Advanced Prompt injection Chains testing must only be performed against systems within your authorised scope. The techniques here involve building conversation context deliberately — ensure all testing conversations are conducted on test accounts and that no production data is accessed as part of the escalation chain. The single-turn injection that fails in three words can…
How to Stay Safe From AI Threats — A Practical Guide for Everyone (2026) | AI Basics Day 5
🤖 AI BASICS FOR BEGINNERS FREE Course Hub → Day 5 of 5 · 🎉 100% complete! Four days ago, "AI" was a fuzzy word you'd heard everywhere but couldn't explain. Today you know what AI actually is, how it learns from examples, the six different types running around in every app you use, and six ways those systems can be attacked. That's real knowledge — not buzzwords. Day 5 is where all of it becomes useful in real life.…
Tuesday, June 9, 2026
How Hackers Attack AI Systems — 6 Real Attack Types Explained Simply (2026) | AI Basics Day 4
🤖 AI BASICS FOR BEGINNERS FREE Course Hub → Day 4 of 5 · 80% complete ⚠️ For Learning Only. Understanding how AI attacks work is how you learn to protect AI systems. All exercises here use systems you're allowed to test. Never try these things on systems without permission. In 2023, a researcher typed a single sentence into a public AI assistant and made it completely ignore all its rules. No hacking tools. No code. No special skills. Just…
Sunday, June 7, 2026
Types of AI Explained — With Real Examples (2026) |AI Basics Day 3
🤖 AI BASICS FOR BEGINNERS FREE Course Hub → Day 3 of 5 · 60% complete Did you know that when you scroll through TikTok, at least three completely different types of AI are working at the same time? One AI recognises what's in each video (computer vision). A second AI predicts which videos you personally want to watch next (recommendation AI). And if there's text or captions in the video, a third AI is reading and understanding them (language…
Subscribe to:
Posts (Atom)