Skip to content
TrustLayerLabs
Get a fixed-price quote in 24h
Back to Home
Technical Reference & Terminology

API & SaaS Security Glossary

Concise engineering definitions of offensive security testing terms, multi-tenant authorization vulnerabilities, AI/LLM attack surfaces, and audit readiness frameworks.

Last updated: · Written by Nagasrinivasa Rao (Lead Security Architect)

Broken Object Level Authorization (BOLA)

An API access-control flaw (OWASP API1) in which an authenticated user can read, modify, or delete records belonging to another user or tenant by changing an object identifier in a request. It typically returns valid HTTP responses with unauthorized data because ownership checks are missing at the data-access layer.

Broken Function Level Authorization (BFLA)

An authorization defect in which lower-privileged accounts or external users can invoke administrative, billing, or internal API functions by targeting controller routes directly. Relying solely on hidden frontend buttons rather than server-side role checks exposes sensitive administrative endpoints.

Insecure Direct Object Reference (IDOR)

A class of access-control vulnerability where an application uses user-supplied identifiers (such as sequential numeric IDs, filenames, or UUIDs) to retrieve database records without verifying session authorization. In APIs, IDOR frequently overlaps with Broken Object Level Authorization (BOLA).

JSON Web Token (JWT)

A compact, self-contained token standard defined by IETF RFC 7519 for transmitting signed claims between parties. Weak signing secrets, improper algorithm handling, missing audience checks, or unvalidated token revocation allow attackers to forge sessions or escalate privileges.

Server-Side Request Forgery (SSRF)

A vulnerability in which an attacker induces a backend service or LLM tool to issue HTTP requests toward unintended internal endpoints, cloud metadata services, or private network hosts. Outbound URL fetchers, webhook callbacks, PDF renderers, and model tool integrations are common SSRF vectors.

Multi-Tenant Isolation

Architectural and runtime enforcement ensuring that users, background workers, caches, and database queries scoped to Tenant A can never access or affect data belonging to Tenant B in shared SaaS infrastructure. Manual multi-tenant testing pairs separate tenant accounts to probe cross-organization boundaries directly.

Vulnerability Assessment and Penetration Testing (VAPT)

A combined security evaluation practice pairing systematic discovery of weaknesses (vulnerability assessment) with active, human-led exploitation of flaws and chained logic paths (penetration testing). VAPT validates whether technical and architectural controls prevent unauthorized access under realistic attack conditions.

Manual Penetration Testing

Security testing performed directly by practitioners who model authentication workflows, user roles, state machines, and business logic rather than relying solely on automated signature scanners. Manual tests produce reproducible proof-of-concept steps that reduce false-positive triage noise for engineering teams.

Prompt Injection (Direct and Indirect)

An adversarial technique targeting Large Language Model (LLM) applications where malicious instructions alter model behavior or override system prompts. Indirect prompt injection embeds instructions inside untrusted retrieved documents, emails, or web pages processed by an autonomous agent.

Retrieval-Augmented Generation (RAG) Poisoning

An attack in which tampered documents or poisoned embeddings enter a vector database so that future retrieval steps surface false guidance, leak cross-tenant context, or trigger hidden agent tool instructions. Tenant scoping on vector collections prevents one customer corpus from bleeding into another.

SOC 2 (System and Organization Controls 2)

An AICPA auditing standard evaluating service organizations against Trust Services Criteria covering Security, Availability, Processing Integrity, Confidentiality, and Privacy. Independent manual penetration testing and retest attestation letters are designed to support SOC 2 Type I and Type II security evidence requests.

ISO/IEC 27001

The international standard specifying requirements for establishing, implementing, maintaining, and continually improving an Information Security Management System (ISMS). Technical vulnerability management and independent penetration testing help validate Annex A operational controls.

Rules of Engagement (RoE)

A written agreement executed prior to security testing defining authorized hosts, API endpoints, testing windows, prohibited techniques, emergency contacts, and escalation protocols. RoE documentation alongside written Authorization-to-Test ensures every engagement stays lawful and inside scope.

security.txt (RFC 9116)

A standardized machine-readable text file hosted at /.well-known/security.txt specifying how external security researchers can report vulnerabilities, which contact channels to use, and where policy details are published.

Need manual verification across these security controls in your stack? Review our API Security Testing and Multi-Tenant SaaS VAPT services or get a fixed-price quote in 24h.

Sources