- GEN AI SECURITY
- Solution landscape
AI Security Landscape – Solutions Directory
The landscape directory includes traditional and emerging security controls addressing Generative
AI, Agentic and AI Red Teaming. It is not a comprehensive list or an endorsement but a community
resource of open source and proprietary solutions.
- Filters:
Landscape
SecOps Stages
- Stage: Govern, Monitor, Operate, Deploy, Release
Company, OSS Project:
Palo Alto Networks
Wiz
Commercial
GenAI LLM
- GenAI LLM
Solution:
Palo Alto Networks AI Runtime Security
Palo Alto Networks AI Runtime Security provides continuous discovery, protection, and monitoring for genAI applications, preventing security risks such as prompt injections, sensitive data leakage, harmful model outputs, and model
Palo Alto Networks AI Runtime Security provides continuous discovery, protection, and monitoring for genAI applications, preventing security risks such as prompt injections, sensitive data leakage, harmful model outputs, and model DoS.
- LLM10:23, LLM09:23, LLM08:23, LLM07:23, LLM06:23, LLM04:23, LLM03:23, LLM02:23, LLM01:23
- Stage: Monitor
Company, OSS Project:
Akto
Wiz
Commercial
GenAI LLM
- GenAI LLM
Solution:
Akto
Akto is an AI agent and MCP Security Solution for Discovery, Red teaming and runtime protection of your agentic Assets ( Agents, MCPs, LLMs). For employees and for homegrown apps.
Akto is an AI agent and MCP Security Solution for Discovery, Red teaming and runtime protection of your agentic Assets ( Agents, MCPs, LLMs). For employees and for homegrown apps.
- Stage: Test & Evaluate
Company, OSS Project:
HackerOne
Wiz
Commercial
Agentic
- Agentic
Solution:
HackerOne AI Red Teaming
Adversarial testing of agentic AI systems using human researchers and autonomous AI agents to identify prompt injection, tool misuse, privilege escalation, and goal manipulation across agent workflows and retrieval pipelines.
Adversarial testing of agentic AI systems using human researchers and autonomous AI agents to identify prompt injection, tool misuse, privilege escalation, and goal manipulation across agent workflows and retrieval pipelines.
- ASI09:26, ASI10:26, ASI01:26, ASI02:26, ASI03:26, ASI05:26, ASI06:26, ASI07:26, ASI08:26
- Stage: Operate, Deploy
Company, OSS Project:
Cisco Systems, Inc.
Wiz
Commercial
GenAI LLM
- GenAI LLM
Solution:
Cisco AI Runtime
Cisco AI Runtime secures GenAI apps to address threats like prompt injections, sensitive data loss, and compliance concerns. Deploy guardrails around safety, privacy, relevancy, and security to govern your AI
Cisco AI Runtime secures GenAI apps to address threats like prompt injections, sensitive data loss, and compliance concerns. Deploy guardrails around safety, privacy, relevancy, and security to govern your AI operations.
- LLM01:23, LLM10:23, LLM09:23, LLM08:23, LLM07:23, LLM06:23, LLM04:23, LLM02:23
- Stage: Govern, Monitor, Deploy, Test & Evaluate
Company, OSS Project:
Humanbound
Wiz
Commercial
Red Teaming
- Red Teaming
Solution:
Humanbound
Continuously test AI agents for security vulnerabilities using adversarial penetration testing. Discover unsanctioned AI services across cloud environments. Track security posture, manage findings, and export guardrails.
Continuously test AI agents for security vulnerabilities using adversarial penetration testing. Discover unsanctioned AI services across cloud environments. Track security posture, manage findings, and export guardrails.
- Stage: Operate
Company, OSS Project:
(Not Provided)
Wiz
Open Source
Red Teaming
- Red Teaming
Solution:
Edictum
Runtime governance library for AI agent tool calls. Deterministic YAML contracts enforce allow/deny/redact verdicts outside the LLM. Structured audit trail, human-in-the-loop approvals, PII redaction. 8 framework adapters. 55µs overhead. MIT
Runtime governance library for AI agent tool calls. Deterministic YAML contracts enforce allow/deny/redact verdicts outside the LLM. Structured audit trail, human-in-the-loop approvals, PII redaction. 8 framework adapters. 55µs overhead. MIT licensed.
- Stage: Monitor
Company, OSS Project:
Brand Engagement Networks
Wiz
Commercial
GenAI LLM
- GenAI LLM
Solution:
SPLX.ai
Red Teaming / Security Testing in the AI CI/CD. The SPLX.ai platform provides continuous testing, guard rail assessments, domain specific test scenarios, AI Inventory which then can be mapped to
Red Teaming / Security Testing in the AI CI/CD. The SPLX.ai platform provides continuous testing, guard rail assessments, domain specific test scenarios, AI Inventory which then can be mapped to security or compliance frameworks such as OWASP Top Ten for LLM Application
- Stage: Test & Evaluate
Company, OSS Project:
Prompt Security
Wiz
Open Source
GenAI LLM
- GenAI LLM
Solution:
Prompt Fuzzer
Prompt Fuzzer is an interactive, open-source tool that empowers developers of GenAI applications to evaluate and enhance the resilience and safety of their system prompts. The users input any system
Prompt Fuzzer is an interactive, open-source tool that empowers developers of GenAI applications to evaluate and enhance the resilience and safety of their system prompts. The users input any system prompt and the relevant configuration, and the Fuzzer starts running its tests.
- LLM01:23, LLM06:23, LLM03:23, LLM02:23
- 361
- 44
- Stage: Operate
Company, OSS Project:
Spellguard.ai
Wiz
Commercial
Agentic
- Agentic
Solution:
Spellguard.ai
Spellguard is a security platform for AI agents that routes all messages and tool calls through a Trusted Execution Environment for real-time policy enforcement, encrypted messaging, and tamper-proof audit logging.
Spellguard is a security platform for AI agents that routes all messages and tool calls through a Trusted Execution Environment for real-time policy enforcement, encrypted messaging, and tamper-proof audit logging.
- ASI01:26, ASI02:26, ASI03:26, ASI05:26, ASI06:26, ASI07:26, ASI08:26, ASI09:26, ASI10:26
- Stage: Govern, Monitor, Operate, Deploy, Test & Evaluate, Augm & Fine Tune Data, Scope & Plan
Company, OSS Project:
Prediction Guard
Wiz
Commercial
GenAI LLM
- GenAI LLM
Solution:
Prediction Guard
Prediction Guard is a security-first AI platform that unifies: infrastructure sovereignty, standards-aligned governance, and no-code agent development into a single, hardened system. AI is a Distributed System, Not a Model.
Prediction Guard is a security-first AI platform that unifies: infrastructure sovereignty, standards-aligned governance, and no-code agent development into a single, hardened system. AI is a Distributed System, Not a Model. Risk emerges from the complex mesh of integrations between models, tools, data sources, and users. Prediction Guard specifically addresses AI governance and system-level security across the ability to deploy, manage and govern AI systems made up of models, tools and applications across distributed cloud architectures.
- LLM04:25, LLM05:25, LLM06:25, LLM07:25, LLM08:25, LLM09:25, LLM10:25, LLM01:25, LLM02:25, LLM03:25
- Stage: Monitor, Operate, Test & Evaluate
Company, OSS Project:
Straiker Inc
Wiz
Commercial
GenAI LLM
- GenAI LLM
Solution:
Straiker AI
Secure AI Applications using two products. Ascend AI provides pentesting/red teaming across all layers of the applications. Defend AI provides visibility, guardrails for AI applications. With both approaches, we take
Secure AI Applications using two products. Ascend AI provides pentesting/red teaming across all layers of the applications. Defend AI provides visibility, guardrails for AI applications. With both approaches, we take a look at the threat vector at the application layer and not just the models
- LLM01:25, LLM02:25, LLM05:25, LLM06:25, LLM07:25, LLM09:25
- Stage: Test & Evaluate
Company, OSS Project:
AIM Intelligence
Wiz
Commercial
GenAI LLM
- GenAI LLM
Solution:
AIM Supervisor
AIM Supervisor integrates AIM RED for automated AI vulnerability testing, AIM GUARD for real-time threat detection and mitigation, and AIM Benchmark for comprehensive safety evaluations, ensuring continuous, adaptive protection against
AIM Supervisor integrates AIM RED for automated AI vulnerability testing, AIM GUARD for real-time threat detection and mitigation, and AIM Benchmark for comprehensive safety evaluations, ensuring continuous, adaptive protection against evolving AI security threats.
- LLM04:25, LLM01:25, LLM02:25, LLM03:25
- Stage: Test & Evaluate
Company, OSS Project:
Eroun&Company
Wiz
Commercial
GenAI LLM
- GenAI LLM
Solution:
SAIFE X RedTeam
RedTeam solution to automate detection of malicious prompt attack vulnerabilities against LLM
RedTeam solution to automate detection of malicious prompt attack vulnerabilities against LLM
- LLM01:25, LLM02:25, LLM03:25, LLM07:25, LLM10:25
- Stage: Deploy
Company, OSS Project:
Pangea
Wiz
Commercial
Agentic
- Agentic
Solution:
AI Security Platform
Pangea's AI Security Platform empowers security teams to protect workforce AI use, enable secure AI innovation, and ship secure AI applications quickly
Pangea's AI Security Platform empowers security teams to protect workforce AI use, enable secure AI innovation, and ship secure AI applications quickly
- ASI01:26, ASI02:26, ASI03:26, ASI05:26, ASI06:26, ASI07:26, ASI08:26, ASI09:26, ASI10:26
- Stage: Develop & Experiment
Company, OSS Project:
Pangea
Wiz
Commercial
GenAI LLM
- GenAI LLM
Solution:
Pangea Data Guard
Prompt inputs, responses, and data ingestion from external sources can all be evaluated for malicious content with Pangea's Data Guard to protect LLMs and users from threatening content.
Prompt inputs, responses, and data ingestion from external sources can all be evaluated for malicious content with Pangea's Data Guard to protect LLMs and users from threatening content.
- LLM10:23, LLM07:23, LLM03:23, LLM02:23
- Stage: Test & Evaluate
Company, OSS Project:
Harmbench
Wiz
Open Source
GenAI LLM
- GenAI LLM
Solution:
Harmbench
HarmBench is a new evaluation framework for automated red teaming and robust refusal.
HarmBench is a new evaluation framework for automated red teaming and robust refusal.
- LLM09:23, LLM08:23, LLM06:23, LLM03:23, LLM02:23, LLM01:23
- 290
- 49
- Stage: Operate
Company, OSS Project:
Dyana
Wiz
Open Source
GenAI LLM
- GenAI LLM
Solution:
Dyana
Dyana is a sandbox environment using Docker and Tracee for loading, running and profiling a wide range of files, including machine learning models, ELF executables, Pickle serialized files, Javascripts and
Dyana is a sandbox environment using Docker and Tracee for loading, running and profiling a wide range of files, including machine learning models, ELF executables, Pickle serialized files, Javascripts and more. It provides detailed insights into GPU memory usage, filesystem interactions, network requests, and security related events.
- LLM03:25, LLM04:25
- 109
- 4
- Stage: Monitor, Operate, Deploy, Test & Evaluate
Company, OSS Project:
HiveTrace
Wiz
Commercial
Agentic
- Agentic
Solution:
HiveTrace
HiveTrace AI monitors gen AI and agentic traffic in real time, inspecting prompts and responses to detect injection attempts, privacy leaks, and hidden system instructions, then issues immediate alerts and
HiveTrace AI monitors gen AI and agentic traffic in real time, inspecting prompts and responses to detect injection attempts, privacy leaks, and hidden system instructions, then issues immediate alerts and logs events for compliance audits.
- ASI01:26, ASI02:26, ASI03:26, ASI05:26, ASI06:26, ASI07:26, ASI08:26, ASI09:26, ASI10:26
- Stage: Govern
Company, OSS Project:
Prompt Security
Wiz
Commercial
GenAI LLM
- GenAI LLM
Solution:
Prompt Security
Prompt Security delivers a complete platform for safeguarding every aspect of Generative AI in the organization. From preventing Shadow AI and data privacy risks associated with the use of GenAI
Prompt Security delivers a complete platform for safeguarding every aspect of Generative AI in the organization. From preventing Shadow AI and data privacy risks associated with the use of GenAI tools and copilots by employees, to protecting from Prompt Injection, Jailbreaks, and other GenAI risks in homegrown applications.
- Stage: Govern
Company, OSS Project:
Cranium
Wiz
Commercial
GenAI LLM
- GenAI LLM
Solution:
Cranium Platform and AI Trust Hub
Whether organizations are builders and/or consumers of AI, Cranium offers a comprehensive platform that enables complete security, compliance, and trust across the entire AI supply chain.
Whether organizations are builders and/or consumers of AI, Cranium offers a comprehensive platform that enables complete security, compliance, and trust across the entire AI supply chain.
- LLM04:25, LLM05:25, LLM06:25, LLM07:25, LLM08:25, LLM09:25, LLM10:25, LLM01:25, LLM02:25, LLM03:25
- Stage: Release
Company, OSS Project:
NeuralTrust
Wiz
Commercial
Agentic
- Agentic
Solution:
NeuralTrust Model Scanner
NeuralTrust Model Scanner inspects AI models to detect vulnerabilities, unsafe code, and corrupted or poisoned artifacts, while verifying model integrity through cryptographic checks and enabling secure, auditable model supply chains.
NeuralTrust Model Scanner inspects AI models to detect vulnerabilities, unsafe code, and corrupted or poisoned artifacts, while verifying model integrity through cryptographic checks and enabling secure, auditable model supply chains.
- ASI05:26, ASI06:26, ASI07:26, ASI08:26, ASI09:26, ASI10:26, ASI01:26, ASI02:26, ASI03:26