Darkhunt AI Security
darkhunt.ai
- Free plan
- Web
- API
- Self-hosted
What Darkhunt AI Security is
Darkhunt AI Security continuously tests, monitors and protects AI agents in production. It discovers models, agents and MCP servers across SaaS, cloud, endpoints and code, including sanctioned and shadow AI. Its observability tools log and replay agent activity and make prompts, responses and tool calls searchable. Automated red teaming tests agents using their real tools, permissions and data access, including for prompt injection, jailbreaks and data exfiltration. The platform also checks decision integrity for issues such as manipulated figures, suppressed alerts, RAG poisoning and confident hallucinations, and traces answers to source data points. Findings can become inline runtime policies, while an MCP gateway proxies requests and responses. Listed integrations include OpenAI, Anthropic, Azure, AWS Bedrock, Gemini and self-hosted systems. Deployment options are cloud, hybrid and self-hosted. Findings and controls can map to frameworks including the EU AI Act, NIST AI RMF, OWASP Top 10 for LLMs, MITRE ATLAS and ISO 42001. The Free plan includes full platform access and community support; paid options are usage-based or custom annual, with prices not listed.
Who it suits
Darkhunt suits teams responsible for AI agents, LLM-powered applications, chatbots or RAG applications in production. It may also suit organizations that need cloud, hybrid or self-hosted deployment and framework mapping.
Good to know, and what to watch
- Discovers sanctioned and shadow AI systems.
- Logs, replays and searches agent activity.
- Automates red-team testing with real permissions.
- Offers cloud, hybrid and self-hosted deployment.
- Free plan includes full platform access.
- Free plan support is community-based.
Plans
Every plan and priceCustom Annual
Pay as you go
Free
Free plan
Getting started
- Open https://darkhunt.ai/
- Choose API, web, or self-hosted access
- Select cloud, hybrid, or self-hosted deployment
- Start with the Free plan at no cost, requiring no card
- Connect listed model providers, frameworks, SIEM, or cloud platforms
Limits to check
The Free plan costs nothing and includes full platform access with community support. Pay as you go has usage-based credits, while Custom Annual uses a fixed annual commitment and volume pricing.
The verdict
Darkhunt combines agent discovery and observability with automated testing and runtime controls. Its free plan provides full platform access, while the listed paid plans do not show prices.
Questions people ask
Is there a free plan?
Yes. Free costs nothing, includes full platform access and community support, and requires no card.
Which deployment options are available?
Darkhunt supports cloud, hybrid, and self-hosted deployment, with API and web access also listed.
What systems can it integrate with?
Listed systems include OpenAI, Anthropic, Azure, AWS Bedrock, Gemini, and self-hosted systems. Pre-built integrations are also listed for agent frameworks, model providers, SIEM, and cloud platforms.
Does it support compliance frameworks?
Findings or controls can map to the EU AI Act, NIST AI RMF, OWASP Top 10 for LLMs, MITRE ATLAS, and ISO 42001.
What support comes with each plan?
Free includes community support. Pay as you go includes dedicated support, while Custom Annual adds dedicated support and onboarding.
What the maker publishes
- Agent inventory
- It discovers models, agents, and MCP servers, including sanctioned and shadow AI across SaaS, cloud, endpoints, and code. source
- Company security
- Darkhunt Limited says it completed a SOC 2 Type I examination covering Security, Availability, and Confidentiality in August 2026, with Type II underway. source
- Decision integrity
- It tests for manipulated figures, suppressed alerts, RAG poisoning, and confident hallucinations, and traces answers to their source data points. source
- Deployment
- Darkhunt says it supports cloud, hybrid, and self-hosted deployment. source
- Framework mapping
- The platform maps findings or controls to the EU AI Act, NIST AI RMF, OWASP Top 10 for LLMs, MITRE ATLAS, and ISO 42001. source
- Integrations
- The site lists OpenAI, Anthropic, Azure, AWS Bedrock, Gemini, and self-hosted systems, and says it offers pre-built integrations for agent frameworks, model providers, SIEM, and cloud platforms. source
- Observability
- It logs, replays, and makes searchable agent prompts, responses, and tool calls. source
- Purpose
- Darkhunt continuously tests, monitors, and protects AI agents in production. source
- Red teaming
- Automated red teaming tests agents with their real tools, permissions, and data access for attacks including prompt injection, jailbreaks, and data exfiltration. source
- Runtime protection
- Findings can become inline runtime policies, and an MCP gateway proxies requests and responses. source
- Support
- The free plan includes community support, pay-as-you-go includes dedicated support, and custom annual includes dedicated support and onboarding. source
- Trust documentation
- The Trust Center lists SOC 2 Type 1 compliance and a SOC 2 Type I report, with access to sensitive documentation available by request. source
Checked against the maker's pages on 28 Sept 2026.