OWASP API Security Top 10 (2023)
The 10 categories in the OWASP API Security Top 10 (2023), each mapped to CWE weaknesses and how TurboPentest tests for it.
- API1:2023Broken Object Level Authorization
Attackers manipulate object IDs in API requests to access objects belonging to other users. The most prevalent and impactful API vulnerability.
- API2:2023Broken Authentication
Authentication mechanisms that are flawed, missing, or implemented incorrectly. Includes weak credential handling, missing rate limits on auth endpoints, JWT validation flaws.
- API3:2023Broken Object Property Level Authorization
Combines the former Excessive Data Exposure (over-returning fields) and Mass Assignment (over-accepting fields) into one category covering field-level authorization failures.
- API4:2023Unrestricted Resource Consumption
APIs without quotas, rate limits, or resource-cost controls allow DoS, financial loss (cloud bill exhaustion), or quota exhaustion attacks.
- API5:2023Broken Function Level Authorization
Attackers can call admin/elevated functions with regular-user credentials by manipulating endpoints or HTTP methods.
- API6:2023Unrestricted Access to Sensitive Business Flows
Business flows (purchase, ticket-buying, comment-posting) that can be abused at scale by automation despite individual requests being authorized.
- API7:2023Server Side Request Forgery (SSRF)
API fetches a remote resource without validating the user-supplied URL, allowing attackers to make the server reach internal/cloud-metadata endpoints.
- API8:2023Security Misconfiguration
Insecure defaults, incomplete configuration, open cloud storage, missing security headers, verbose errors, unpatched systems.
- API9:2023Improper Inventory Management
Old API versions, undocumented endpoints, debug/staging APIs left exposed alongside production. Inventory drift creates a wider attack surface than the documented API.
- API10:2023Unsafe Consumption of APIs
Trusting third-party API responses without validation. Compromised partner APIs become a vector into the consuming application.
Find these issues before an attacker does
TurboPentest runs an agentic AI pentest against your target and reports findings with proof, from $99 per target.
Start a pentest