AGENT SKILL

testing-api-for-broken-object-level-authorization

Tests REST and GraphQL APIs for Broken Object Level Authorization (BOLA/IDOR, OWASP API1:2023) by intercepting API calls, identifying object ID parameters (numeric IDs, UUIDs, slugs), and systematically substituting IDs belonging to other users to check whether the server enforces per-object authorization. Use when asked to test BOLA or IDOR in an API, verify object-level authorization, or assess an API for access control bypass.

License 许可
Apache-2.0
最近更新
2026-08-25
Stars
★ 0
Forks
0

01 前言 Front Matter

name
testing-api-for-broken-object-level-authorization
tags
[ "api-security", "owasp", "bola", "idor", "authorization", "rest-security" ]
author
mahipal
domain
cybersecurity
license
Apache-2.0
version
1.0.0
nist_csf
[ "PR.PS-01", "ID.RA-01", "PR.DS-10", "DE.CM-01" ]
subdomain
api-security
description
Tests REST and GraphQL APIs for Broken Object Level Authorization (BOLA/IDOR, OWASP API1:2023) by intercepting API calls, identifying object ID parameters (numeric IDs, UUIDs, slugs), and systematically substituting IDs belonging to other users to check whether the server enforces per-object authorization. Use when asked to test BOLA or IDOR in an API, verify object-level authorization, or assess an API for access control bypass.
mitre_attack
[ "T1190", "T1059.007", "T1552.001", "T1027", "T1070" ]

02 资源文件 3

03 SKILL.md 正文

Testing API for Broken Object Level Authorization

When to Use

  • Assessing REST or GraphQL APIs that use object identifiers in URL paths, query parameters, or request bodies
  • Performing OWASP API Security Top 10 assessments where API1:2023 (BOLA) must be tested
  • Testing multi-tenant SaaS applications where users from different tenants should not access each other's data
  • Validating that API endpoints enforce per-object authorization checks beyond just authentication
  • Evaluating APIs after new endpoints are added to ensure authorization middleware is applied consistently

Do not use without written authorization from the API owner. BOLA testing involves accessing or attempting to access other users' data, which requires explicit permission.

Prerequisites

  • Written authorization specifying the target API endpoints and scope of testing
  • At least two test accounts with different privilege levels and distinct data sets
  • Burp Suite Professional or OWASP ZAP configured as an intercepting proxy
  • Authentication tokens (JWT, session cookies, API keys) for each test account
  • API documentation (OpenAPI/Swagger spec) or access to enumerate endpoints
  • Python 3.10+ with requests library for scripted testing
  • Autorize Burp extension installed for automated BOLA detection

Workflow

Step 1: API Endpoint Discovery and Object ID Mapping

Enumerate all API endpoints and identify parameters that reference objects:

From OpenAPI/Swagger Specification:

# Download and parse the OpenAPI spec
curl -s https://target-api.example.com/api/docs/swagger.json | python3 -m json.tool

# Extract all endpoints with path parameters
curl -s https://target-api.example.com/api/docs/swagger.json | \
  python3 -c "
import json, sys
spec = json.load(sys.stdin)
for path, methods in spec.get('paths', {}).items():
    for method, details in methods.items():
        if method in ('get','post','put','patch','delete'):
            params = [p['name'] for p in details.get('parameters',[]) if p.get('in') in ('path','query')]
            if params:
                print(f'{method.upper()} {path} -> params: {params}')
"

From Burp Suite Traffic:

  1. Browse the application as User A, exercising all features that involve data creation and retrieval
  2. In Burp, go to Target > Site Map and filter for API paths (e.g., /api/v1/, /graphql)
  3. Look for patterns: /api/v1/users/{id}, /api/v1/orders/{order_id}, /api/v1/documents/{doc_uuid}
  4. Note the object ID format: sequential integers (predictable), UUIDs (less predictable), or encoded values

Classify Object ID Types:

ID Type Example Predictability BOLA Risk
Sequential Integer /orders/1042 High - increment/decrement Critical
UUID v4 /orders/550e8400-e29b-41d4-a716-446655440000 Low - random Medium (if leaked)
Encoded/Hashed /orders/base64encodedvalue Medium - decode and predict High
Composite /users/42/orders/1042 High - multiple IDs to swap Critical
Slug /profiles/john-doe Medium - guess usernames High

Step 2: Baseline Request Capture with Authenticated User

Capture legitimate requests for User A and User B:

import requests

BASE_URL = "https://target-api.example.com/api/v1"

# User A credentials
user_a_token = "Bearer eyJhbGciOiJSUzI1NiIsInR5cCI6IkpXVCJ9..."
user_a_headers = {"Authorization": user_a_token, "Content-Type": "application/json"}

# User B credentials
user_b_token = "Bearer eyJhbGciOiJSUzI1NiIsInR5cCI6IkpXVCJ9..."
user_b_headers = {"Authorization": user_b_token, "Content-Type": "application/json"}

# Step 1: Identify User A's objects
user_a_profile = requests.get(f"{BASE_URL}/users/me", headers=user_a_headers)
user_a_id = user_a_profile.json()["id"]  # e.g., 1001

user_a_orders = requests.get(f"{BASE_URL}/users/{user_a_id}/orders", headers=user_a_headers)
user_a_order_ids = [o["id"] for o in user_a_orders.json()["orders"]]  # e.g., [5001, 5002]

# Step 2: Identify User B's objects
user_b_profile = requests.get(f"{BASE_URL}/users/me", headers=user_b_headers)
user_b_id = user_b_profile.json()["id"]  # e.g., 1002

user_b_orders = requests.get(f"{BASE_URL}/users/{user_b_id}/orders", headers=user_b_headers)
user_b_order_ids = [o["id"] for o in user_b_orders.json()["orders"]]  # e.g., [5003, 5004]

print(f"User A (ID: {user_a_id}): Orders {user_a_order_ids}")
print(f"User B (ID: {user_b_id}): Orders {user_b_order_ids}")

Step 3: BOLA Testing - Horizontal Privilege Escalation

Attempt to access User B's objects using User A's authentication:

import json

results = []

# Test 1: Access User B's profile with User A's token
resp = requests.get(f"{BASE_URL}/users/{user_b_id}", headers=user_a_headers)
results.append({
    "test": "Access other user profile",
    "endpoint": f"GET /users/{user_b_id}",
    "auth": "User A",
    "status": resp.status_code,
    "vulnerable": resp.status_code == 200,
    "data_leaked": list(resp.json().keys()) if resp.status_code == 200 else None
})

# Test 2: Access User B's orders with User A's token
for order_id in user_b_order_ids:
    resp = requests.get(f"{BASE_URL}/orders/{order_id}", headers=user_a_headers)
    results.append({
        "test": f"Access other user order {order_id}",
        "endpoint": f"GET /orders/{order_id}",
        "auth": "User A",
        "status": resp.status_code,
        "vulnerable": resp.status_code == 200
    })

# Test 3: Modify User B's order with User A's token
resp = requests.patch(
    f"{BASE_URL}/orders/{user_b_order_ids[0]}",
    headers=user_a_headers,
    json={"status": "cancelled"}
)
results.append({
    "test": "Modify other user order",
    "endpoint": f"PATCH /orders/{user_b_order_ids[0]}",
    "auth": "User A",
    "status": resp.status_code,
    "vulnerable": resp.status_code in (200, 204)
})

# Test 4: Delete User B's resource with User A's token
resp = requests.delete(f"{BASE_URL}/orders/{user_b_order_ids[0]}", headers=user_a_headers)
results.append({
    "test": "Delete other user order",
    "endpoint": f"DELETE /orders/{user_b_order_ids[0]}",
    "auth": "User A",
    "status": resp.status_code,
    "vulnerable": resp.status_code in (200, 204)
})

# Print results
for r in results:
    status = "VULNERABLE" if r["vulnerable"] else "SECURE"
    print(f"[{status}] {r['test']}: {r['endpoint']} -> HTTP {r['status']}")

Step 4: Advanced BOLA Techniques

Test for less obvious BOLA patterns:

# Technique 1: Parameter pollution - send both IDs
resp = requests.get(
    f"{BASE_URL}/orders/{user_a_order_ids[0]}?order_id={user_b_order_ids[0]}",
    headers=user_a_headers
)
print(f"Parameter pollution: {resp.status_code}")

# Technique 2: JSON body object ID override
resp = requests.post(
    f"{BASE_URL}/orders/details",
    headers=user_a_headers,
    json={"order_id": user_b_order_ids[0]}
)
print(f"Body ID override: {resp.status_code}")

# Technique 3: Array of IDs - include other user's IDs in batch request
resp = requests.post(
    f"{BASE_URL}/orders/batch",
    headers=user_a_headers,
    json={"order_ids": user_a_order_ids + user_b_order_ids}
)
print(f"Batch ID inclusion: {resp.status_code}, returned {len(resp.json().get('orders',[]))} orders")

# Technique 4: Numeric ID manipulation for sequential IDs
for offset in range(-5, 6):
    test_id = user_a_order_ids[0] + offset
    if test_id not in user_a_order_ids:
        resp = requests.get(f"{BASE_URL}/orders/{test_id}", headers=user_a_headers)
        if resp.status_code == 200:
            owner = resp.json().get("user_id", "unknown")
            if str(owner) != str(user_a_id):
                print(f"BOLA: Order {test_id} belongs to user {owner}, accessible by User A")

# Technique 5: Swap object ID in nested resource paths
resp = requests.get(
    f"{BASE_URL}/users/{user_b_id}/orders/{user_b_order_ids[0]}/invoice",
    headers=user_a_headers
)
print(f"Nested resource BOLA: {resp.status_code}")

# Technique 6: Method switching - GET may be blocked but PUT allowed
for method in ['GET', 'PUT', 'PATCH', 'DELETE', 'HEAD', 'OPTIONS']:
    resp = requests.request(
        method,
        f"{BASE_URL}/users/{user_b_id}/settings",
        headers=user_a_headers,
        json={"notifications": False} if method in ('PUT', 'PATCH') else None
    )
    if resp.status_code not in (401, 403, 405):
        print(f"Method {method} on other user settings: {resp.status_code}")

Step 5: Automated BOLA Detection with Autorize (Burp Suite)

Configure Autorize for automated detection:

  1. Install Autorize from the BApp Store in Burp Suite Professional
  2. In the Autorize tab, paste User B's authentication cookie or header
  3. Configure the interception filters:
    • Include: .*\/api\/.* (only API paths)
    • Exclude: .*\.(js|css|png|jpg)$ (skip static assets)
  4. Set the enforcement detector:
    • Add conditions where response length or status code differs between User A and User B
    • Mark as "enforced" if User A gets 403/401 for User B's resources
    • Mark as "bypassed" if User A gets 200 with User B's data
  5. Browse the application as User A; Autorize automatically replays each request with User B's token
  6. Review the Autorize results table:
    • Green = Authorization enforced (secure)
    • Red = Authorization bypassed (BOLA vulnerability)
    • Orange = Needs manual review (ambiguous response)

Step 6: GraphQL BOLA Testing

# Test BOLA in GraphQL queries using node/ID relay pattern
# User A queries User B's order by global relay ID
query {
  node(id: "T3JkZXI6NTAwMw==") {  # Base64 of "Order:5003" (User B's)
    ... on Order {
      id
      totalAmount
      shippingAddress {
        street
        city
      }
      items {
        productName
        quantity
      }
    }
  }
}

# Test nested object access through relationships
query {
  user(id: "1002") {  # User B's ID
    email
    phoneNumber
    orders {
      edges {
        node {
          id
          totalAmount
          paymentMethod {
            lastFourDigits
          }
        }
      }
    }
  }
}

Key Concepts

Term Definition
BOLA Broken Object Level Authorization (OWASP API1:2023) - the API does not verify that the authenticated user has permission to access the specific object referenced by the request
IDOR Insecure Direct Object Reference - a closely related term where the application uses user-controllable input to directly access objects without authorization checks
Horizontal Privilege Escalation Accessing resources belonging to another user at the same privilege level by manipulating object identifiers
Vertical Privilege Escalation Accessing resources or functions restricted to a higher privilege level (e.g., regular user accessing admin endpoints)
Object ID Enumeration Predicting valid object identifiers by analyzing their format (sequential integers, UUID patterns, encoded values)
Autorize A Burp Suite extension that automates authorization testing by replaying requests with different user tokens

Tools & Systems

  • Burp Suite Professional: Intercepting proxy for capturing and manipulating API requests with Autorize extension for automated BOLA testing
  • OWASP ZAP: Open-source alternative with Access Control Testing add-on for authorization boundary testing
  • Autorize: Burp extension that automatically detects authorization enforcement by replaying requests with different user contexts
  • Postman: API testing platform for crafting and replaying requests with different authentication tokens across collections
  • ffuf: Web fuzzer that can enumerate object IDs at scale: ffuf -u https://api.example.com/orders/FUZZ -w ids.txt -H "Authorization: Bearer token"

Common Scenarios

Scenario: E-Commerce API BOLA Assessment

Context: An e-commerce platform exposes a REST API for its mobile app. The API uses sequential integer IDs for orders, users, and addresses. Two test accounts are provided: a regular customer (User A, ID 1001) and another customer (User B, ID 1002).

Approach:

  1. Map all endpoints from the Swagger spec at /api/docs: identify 47 endpoints, 23 of which take object IDs
  2. Capture User A's requests for their own resources: profile, orders, addresses, payment methods, wishlist
  3. Replace User A's object IDs with User B's IDs systematically across all 23 endpoints
  4. Find that GET /api/v1/orders/{id} returns any order regardless of ownership (BOLA on read)
  5. Find that PATCH /api/v1/addresses/{id} allows modifying any user's address (BOLA on write)
  6. Find that GET /api/v1/users/{id}/payment-methods leaks payment card last-four digits for any user
  7. Test batch endpoint POST /api/v1/orders/export - accepts array of order IDs and exports all without ownership check
  8. Verify that DELETE /api/v1/orders/{id} correctly returns 403 for non-owned orders (authorization enforced)

Pitfalls:

  • Only testing GET requests and missing BOLA in PUT/PATCH/DELETE methods that allow data modification or destruction
  • Assuming UUIDs prevent BOLA - UUIDs are less predictable but can be leaked in API responses, logs, or URL parameters
  • Not testing nested resource paths where authorization may be checked on the parent but not the child resource
  • Missing BOLA in bulk/batch endpoints that accept arrays of object IDs
  • Not considering that different API versions (v1 vs v2) may have different authorization implementations

Output Format

## Finding: Broken Object Level Authorization in Order API

**ID**: API-BOLA-001
**Severity**: High (CVSS 7.5)
**OWASP API**: API1:2023 - Broken Object Level Authorization
**Affected Endpoints**:
  - GET /api/v1/orders/{id}
  - PATCH /api/v1/addresses/{id}
  - GET /api/v1/users/{id}/payment-methods
  - POST /api/v1/orders/export

**Description**:
The API does not enforce object-level authorization on order retrieval,
address modification, payment method viewing, or order export endpoints.
An authenticated user can access or modify any other user's resources by
substituting object IDs in the request. Sequential integer IDs make
enumeration trivial.

**Proof of Concept**:
1. Authenticate as User A (ID 1001): POST /api/v1/auth/login
2. Retrieve User A's order: GET /api/v1/orders/5001 -> 200 OK (legitimate)
3. Access User B's order: GET /api/v1/orders/5003 -> 200 OK (BOLA - returns full order details)
4. Modify User B's address: PATCH /api/v1/addresses/2002 -> 200 OK (BOLA - address changed)

**Impact**:
- Read access to all 850,000+ customer orders including shipping addresses and order contents
- Write access to any customer's delivery address, enabling package redirection
- Exposure of partial payment card data for all customers

**Remediation**:
1. Implement object-level authorization middleware that verifies the authenticated user owns the requested resource
2. Use authorization checks at the data access layer: `WHERE order.user_id = authenticated_user.id`
3. Replace sequential integer IDs with UUIDs to reduce predictability (defense in depth, not a fix alone)
4. Add authorization tests to the CI/CD pipeline for every endpoint that accepts object IDs
5. Implement rate limiting per user to slow enumeration attempts

04 同仓库更多 Skills

vikasvardhanv
vikasvardhanv/jerry_skills

syllabus

Generates a curated supplementary reading list from any course syllabus using Consensus academic search. Grill-me intake (syllabus input format + course audience + year range) plus a grouping forcing-options checkpoint before any search runs — so the reading list matches the course's level and recency need. Parses the syllabus to extract topics and learning outcomes, searches Consensus for recent peer-reviewed papers per topic, and produces a professionally formatted .docx with clickable Consensus links, plain-language summaries calibrated to audience level, and Bloom-higher-order discussion questions tied to course learning goals. Use when the user uploads a syllabus, course outline, or curriculum document and wants supplementary readings (e.g., 'create a reading list from this syllabus', 'find recent papers for my course') — even casual mentions with a syllabus attached should trigger this skill.

★ 0
vikasvardhanv
vikasvardhanv/jerry_skills

pulse

Multi-source recency research skill that takes the pulse of any topic across Reddit, Hacker News, the open web, and optionally X/Twitter within a configurable recent window (default 30 days). Forcing intake clarifies topic specificity, angle (trend/sentiment/problems/opportunities/comparison), time window, and platform scope before searching. Returns a synthesized briefing with citations, engagement metrics, and cross-platform pattern analysis. Use when the user requests multi-source recency intelligence on a topic (e.g., 'pulse on [topic]', 'what's happening with [topic]', 'what are people saying about [topic]', 'current conversation about [topic]', 'take the pulse of [topic]', 'trending: [topic]', 'find me info on [topic]'), and for competitor research, trend discovery, tool comparisons, and audience sentiment analysis.

★ 0
vikasvardhanv
vikasvardhanv/jerry_skills

patent

Patent prior-art and landscape intelligence skill — not generic patent help. Commits to one of five sub-use-cases via forcing intake (novelty search / freedom-to-operate / competitive landscape / acquisition diligence / litigation prior-art) before any search runs. Searches Google Patents, Espacenet, USPTO, and optionally Lens.org for citation-graph signals. Output is an editable Word document (.docx) with verdict, ranked closest art (claim-text extracted), CPC-class-aware landscape, family-resolved hits, geographic coverage, FTO flags where applicable, strategy recommendations, and full audit log. Use when the user asks for patent searching or analysis (e.g., 'prior art search for [invention]', 'freedom to operate analysis for [product]'). Produces search signal, not legal advice — always recommends consulting a patent attorney before filing or licensing decisions. Trademark, copyright, and trade-secret questions are out of scope.

★ 0
vikasvardhanv
vikasvardhanv/jerry_skills

notebooklm

Browser automation skill for controlling Google's NotebookLM. Use when the user wants anything done in NotebookLM (e.g., 'open NotebookLM', 'check my [name] notebook', 'ask my notebook about X', 'add [source] to NotebookLM', 'generate a Video Overview from my notebook', 'use NotebookLM Studio'). Handles reading and querying notebooks, adding sources (URLs, text, files, YouTube links, synthesized content), generating Studio outputs (Audio/Video Overviews, Mind Maps, Reports incl. Briefing Doc/Study Guide/FAQ, Flashcards, Quiz, slide decks, infographics — discover the exact set from the live Studio panel; the UI evolves fast), and creating new notebooks. Requires browser automation environment — fails gracefully when unavailable.

★ 0
vikasvardhanv
vikasvardhanv/jerry_skills

litreview

Academic literature orientation skill that searches papers via free keyless APIs (PubMed E-utilities + OpenAlex) by default — with the Consensus MCP as an optional enhancement lane when connected — builds a strategic search plan using PICO (default) or SPIDER / Decomposition / hybrid as fallbacks, and synthesizes findings into a formatted Word (.docx) research guide. Grill-me intake (research question specificity + framework hint + tentative depth) before the recon search; a second forcing checkpoint after Phase 2 confirms framework + sub-areas + depth before searches consume budget. Configurable depth (5/10/20 queries) controls coverage vs. speed. Output is a 'launching pad' — an orientation guide that lets a researcher dive in confidently, not a finished review. Use when the user starts literature-oriented research (e.g., 'litreview on [topic]', 'literature review on [topic]', 'I'm starting a literature review on X', 'I'm writing a paper on X', 'help me research X', 'I'm doing research on X', 'can you help me research X'). Do NOT use for single one-off paper searches wanting a quick list — that's a plain PubMed/OpenAlex (or Consensus) query.

★ 0
vikasvardhanv
vikasvardhanv/jerry_skills

grants

NIH grant research skill for clinical researchers. Grill-me intake (research idea + career stage + preliminary data + environment + submission posture + known institute targets) locks down the funding strategy before any search runs. Runs a 5-facet Consensus positioning analysis (with draft Significance/Innovation language), maps the research to the right NIH institutes and study sections via RePORTER, finds NOSIs and funded overlap, and produces an editable Word document (.docx) with budget/scope-aware mechanism recommendations, submission timelines, and a mandatory program officer recommendation. Use when the user asks about research funding or makes any grant-related request (e.g., 'grants for [topic]', 'find grants for my research idea', 'what grants match my research', 'help me find NIH funding', 'grant opportunities for my research'). NIH-only scope — non-NIH funders (PCORI, DOD CDMRP, VA, foundations) are out of scope and flagged at intake.

★ 0