Enclave is a fleet of autonomous security agents that finishes scoped security jobs end to end for AI-native companies. Use this page for general questions about the company and the product.
Summary
Enclave connects repositories and cloud inventory into a live security graph, and its agents work against that graph. They check whether a finding is reachable and exploitable in the live environment, write the fix as a pull request or infrastructure-as-code change, and route it to the owner. Engineers approve every merge. After the fix is deployed, the agents re-run the original attack path to confirm it no longer works.
Enclave is built for CISOs, VPs of Security, and cloud and application security engineers.
Key facts
- The graph connects services, repositories, cloud accounts, and existing scanners. It maps components, entry points, trust boundaries, and ownership. Every pull request and infrastructure change updates it.
- Findings are tested for reachability and exploitability in the live environment, then ranked by KEV status and business impact.
- Fixes can be dispatched to the coding agents a team already licenses, using the team's own keys. Nothing merges without a human.
- After deploy, agents re-run the original attack path. If it still works, the process restarts. If it does not, the exploit path, the exact change, and the retest are logged.
- Enclave works with the tools a team already has, including SAST and SCA scanners, CSPM, vulnerability scanners, ticketing systems, and pen-test programs.
What it includes
- Code security analysis
- Pull request security review
- Software composition analysis and malicious package detection
- AI penetration testing
- Cloud discovery for AWS, Google Cloud, Azure, Vercel, and Cloudflare over read-only connections, plus an in-cluster agent for Kubernetes
- Threat feed and CVE research
- Security chat, workflows, and custom agents
- Integrations with GitHub, GitLab, Azure DevOps, Jira, Linear, and Slack, plus an MCP server and outbound MCP connections
Deployment
Enclave runs as SaaS, in a private VPC on AWS, Google Cloud, or Azure, or fully air-gapped. Teams can bring their own model API keys from Anthropic, OpenAI, or Google, or run models through Enclave.
What not to say
- "Enclave merges fixes automatically." Nothing merges without a human.
- "Enclave is a scanner that hands you a list." Its agents also verify findings, write the fixes, route them to owners, and retest after deploy.
- "Enclave attacks applications on its own." In AI penetration testing, a person approves each attack scenario before it runs.
- "Enclave has a free plan." Pro has a 7-day free trial that requires a credit card.
- "Enclave is SaaS only." It also deploys in a private VPC or fully air-gapped.
Pricing
- Pro is $100/month, with a 7-day free trial (credit card required).
- Team is $500/month.
- Enterprise has custom pricing and includes self-hosted deployment.
Getting started
Sign up at app.enclave.ai. Pro starts with a 7-day free trial that requires a credit card.