See Wallarm's AI Control Platform In Action
Talk to an Expert
Wallarm API Security brings API discovery, API and web protection, abuse prevention, and MCP controls into one product. Imperva, now part of Thales, delivers the capabilities compared here across Imperva Cloud WAF, WAF Gateway, Imperva API Security, and the Thales AI Security Fabric.
Integrated API security
for API discovery, WAAP, and MCP.
Wallarm
✓ Yes
Imperva
– Across multiple products
API leak detection
Exposed credentials on public sources
Wallarm
✓ Yes
Imperva
✕ No [Verify]
Every claim cites public documentation and names the Imperva product it applies to.
See Wallarm In Action
Ten capabilities, compared product by product and cited to public documentation. Claims about Imperva come from Imperva's and Thales's own public materials.
✓
SUPPORTED
–
Partial or indirect
✕
Not supported or not documented
Capability
Wallarm API Security
Imperva Cloud WAF
Imperva WAF Gateway
Imperva API Security
Thales AI Security Fabric
Inline enforcement
✓
Request-level blocking, session blocking, IP controls [3]
✓
Inline cloud reverse proxy via DNS [N1][Verify]
–
Not documented
API protocols
✓
REST, GraphQL, SOAP, gRPC payloads [N4][Verify: 2020 source]
Not documented
✓
REST, GraphQL, gRPC, SOAP; WebSocket not listed [N5]
n/a
Behavioral API detection
✓
BOLA, enumeration, forced browsing, brute force [5]
–
Enforces API Security BOLA verdicts [N3]
–
Enforces API Security BOLA verdicts [N3]
n/a
Credential stuffing
✓
850M+ HIBP-derived records [6]
–
Account Takeover Protection add-on [N6]
–
Legacy ThreatRadar ATO; status unclear [N7][Verify]
n/a
n/a
Bot and automated abuse
✓
ATO, scraping, IP and session rotation [7]
–
Advanced Bot Protection, separate product [N8]
–
Pairs with Advanced Bot Protection [N9]
n/a
API discovery
✓
Continuous API and MCP inventory from live traffic [4]
–
Via API Security add-on [N9]
–
Via API Security sensor [N10]
n/a
API leak detection
✓
Scans public sources for exposed keys and tokens [8]
✕
Not found
✕
Not found
✕
Not found; token risks cover observed traffic only [N11]
✕
Not found
AI / LLM runtime protection
✓
AI payload inspection at configured request and response points [11]
n/a
n/a
n/a
Wallarm API Security runs one node that detects and blocks in the same engine, whether it sits on-prem, in any cloud, hybrid, or as Kubernetes ingress. Imperva delivers the capabilities compared here through a cloud proxy, an on-prem gateway, and an API security product that can be managed from either side.
✓
Inline deployment handles attacks at the request level, with session and IP-based blocking controls. [3]
✓
The same node runs on-prem, in any cloud, hybrid, or as Kubernetes ingress, so internal east-west APIs are in scope. [13]
✓
Out-of-band modes record malicious traffic for visibility. Inline deployment is required for blocking. [12]
–
Deeper AI-workload governance lives in Wallarm AI Hypervisor, a separate AWS product that deploys on Amazon EKS. [15]
✓
Imperva Cloud WAF routes traffic through Imperva's network for WAF, DDoS, and CDN.
✓
Imperva WAF Gateway runs on customer infrastructure, and Imperva promotes Kubernetes deployment with sidecar sensors and an inline WAF gateway. [17]
Wallarm documents dedicated parsers for JWT, ASP.NET ViewState, Base64, and JSON, so an attack inside a nested format is inspected, not just the outer wrapper. Its behavioral controls run on API sessions rather than single requests.
✓
Dedicated parsers for JWT header and payload, ASP.NET ViewState, Base64, and JSON. [14]
✓
BOLA, enumeration, forced-browsing, and brute-force controls operate on API sessions. [5]
✓
Credential stuffing detection checks passwords against HIBP-derived compromised records. [6]
–
Abuse is blocked once confidence is high enough not to break legitimate traffic, so not every automated request is stopped on first sight. [7]
✓
Imperva API Security builds behavioral baselines for API usage and flags deviations. [17]
✓
BOLA detection and response was added to Imperva Application Security in December 2025. [21]
–
Bot and credential-abuse detection comes from Advanced Bot Protection, integrated with API Security. [19]
–
Parsing depth for nested encodings is not described in the materials reviewed.
These are the request shapes to try, in monitoring mode, against your own deployment. Results depend on your configuration.
A Base64 payload nested inside XML
Two stacked encoding layers
An encoded GraphQL pagination cursor
Parsing GraphQL first, then decoding the cursor
A .NET __VIEWSTATE field
No readable markup to pattern-match against
A payload in the JWT header
The header itself, decoded and inspected, not just the bearer envelope
An archive compressed inside a JSON parameter
Decompressing the parameter to reach the payload
An injection preceded by several hundred KB of filler
Take an injection your WAF blocks unpadded, pad it, and send it again
The same parameter name twice
Whether the engine recombines duplicates the way your application server will
Wallarm detects and blocks in the same engine and can stop a specific API session rather than an entire IP address. Imperva enforces through its WAF products, with API Security supplying discovery and detection.
✓
Session denylisting and blocking act on a specific API session, not only on an IP address. [3]
✓
BOLA, enumeration, forced-browsing, and brute-force controls can run in monitor, IP-block, or session-block mode. [5]
✓
Out-of-band deployment is for visibility; inline deployment is required for blocking. [12]
Wallarm API Security groups MCP traffic into dedicated sessions and enforces tool calls with ACLs by MCP method and primitive, request verification, and validation of tools/call arguments against schemas learned from tools/list. Imperva inspects agent-to-MCP traffic through its Google agent gateway partnership, and Thales has announced an MCP security gateway for 2026.
✓
MCP servers are discovered into the API inventory from live traffic. [4]
✓
MCP sessions group tool calls, resource reads, and prompts, and show the server, method, primitive, attacks, arguments, and request sequence. [9]
✓
ACL policies by MCP method and primitive, request verification, and schema validation of tools/call arguments, with invalid calls blocked. [10]
–
Deeper AI-workload runtime governance lives in Wallarm AI Hypervisor, available on AWS and deployed on Amazon EKS with a separate onboarding flow. [15]
✓
Imperva inspects agent-to-MCP traffic for unauthorized tool execution, anomalous tool calls, and malicious parameter injection via Imperva for Google Cloud, in controlled availability. [20]
–
An MCP security gateway is among the Thales AI Security Fabric capabilities announced for 2026. [22]
–
Thales AI Security Fabric provides runtime security for agentic AI and LLM applications. [22]
Choose based on where enforcement needs to live and how many products you want to operate. If you want discovery, protection, abuse prevention, and MCP controls in one product running inside your own walls, Wallarm fits that shape.
✓
You want API discovery, API and web protection, abuse prevention, and MCP controls in one product and one console. [1][2]
✓
Inspection must run inside your own datacenter or cloud account, including internal east-west APIs. [13]
✓
You need to find exposed API credentials, keys, and tokens that never cross an enforcement point. [8]
✓
You need MCP tool calls validated against learned schemas and ACLs. [10]
✓
You need session-level blocking, not only IP blocking. [3]
✓
Imperva's cloud edge is already your CDN and DDoS layer.
✓
You're consolidating on the broader Thales security portfolio.
✓
Your agents run through Google's agent gateway and you want inspection native to it. [20]
— Raiffeisen Bank
Decided Wallarm fits? You don't have to rip anything out. Layer it alongside Imperva or consolidate at renewal.
Which Imperva product handles API security?
Can Imperva API Security run on-premises?
Yes. Imperva offers a self-managed console that keeps API discovery, classification, and risk assessment on customer premises. [16]
Does Imperva detect BOLA?
How does Imperva secure MCP servers?
Does Imperva detect leaked API keys?
In the public Imperva materials we reviewed, we didn't find a capability for discovering exposed API credentials on public sources. Wallarm AASM scans public resources for exposed credentials, API keys, client secrets, and authorization tokens. [8]
How does Wallarm secure MCP servers?
Is Imperva part of Thales?
Yes. Imperva is a Thales company, and its application security products are marketed as the Imperva Application Security platform. [21]
Sources
Every numbered reference on this page links to public documentation from Wallarm, Imperva, or Thales.
[16]
Imperva — Deploy API Security On-Premises with New Imperva API Security Anywhere Self-Managed Option
https://www.imperva.com/blog/deploy-api-security-on-premises-with-new-imperva-api-security-anywhere-self-managed-option/
[17]
Imperva — On-Premises API Security on Kubernetes
https://www.imperva.com/blog/on-premises-api-security-kubernetes/
[18]
Imperva — API Security and Protection
https://www.imperva.com/resources/resource-library/reports/top-10-things-software-engineering-leaders-need-to-know-about-apis/
[19]
Imperva — API Security Data Sheet V5 (PDF)
https://imperva.com/resources/datasheets/Imperva-API-Security-DS_V5.pdf
[20]
Thales — Securing the Next Generation of AI: Thales and Google Partner to Protect the AI Agent Ecosystem (May 2026)
https://cpl.thalesgroup.com/blog/cybersecurity/thales-google-protect-ai-agent-ecosystem
[21]
Thales via Business Wire — BOLA detection and response in Imperva Application Security (December 2025)
https://www.businesswire.com/news/home/20251211984690/fr
[22]
Thales via Business Wire — Thales AI Security Fabric launch (December 2025)
https://www.businesswire.com/news/home/20251211485558/es
Deploy alongside your existing Imperva estate and see what each engine detects.