Kontext Raises $4M to Stop AI Agents From Going Rogue at Work
Kontext raises $4 million led by 42CAP, with backing from a16z CSX and HTGF, to expand its team and build runtime security for AI agents.
2026-09-24 · 5 min read
Control agent and user behavior in real time.
Kontext is the runtime security platform for your AI workforce. Discover agents, map what they can access, block prohibited actions, detect risk in agent behavior, and manage spend to reduce costs. Keep evidence for every policy decision.
Control agent and user behavior in real time.
Kontext is the runtime security platform for your AI workforce. Discover agents, map what they can access, block prohibited actions, detect risk in agent behavior, and manage spend to reduce costs. Keep evidence for every policy decision.
Kontext protects your AI workforce, from coding assistants and workplace agents to custom-built agents, across endpoints and cloud environments. Bring agent discovery, access policies, risk monitoring, and spend management together across the agents your organization uses.
How it works
Kontext runs in the agent’s environment. Policies evaluate actions at pre-action control points and determine whether to allow or deny them. Risk detection flags activity for investigation, and decision records preserve the policy decision and available outcome. Risk assessments and policy enforcement are distinct controls.
Control agents wherever they run
Discover coding assistants, workplace agents, and custom-built agents across endpoints and cloud environments. Map their access, enforce action policies, monitor risk, and manage spend.
Start in observe mode, enforce when ready
Review the actions a policy would block without interrupting the agent. When the matches look right, enforce the policy at the configured control point. Review decision evidence to validate controls and investigate activity.
Start with visibility
Sign up and create or join a workspace before connecting an endpoint. Endpoint setup requires workspace owner or admin access.
The self-serve quickstart uses macOS and Homebrew. Install with brew install kontext-security/tap/kontext, open Get started in the dashboard for your personal API key, then run kontext setup and paste the key at the prompt. Run kontext doctor to check the installation. Trust the Kontext hooks in Codex when prompted. Follow the quickstart to send a read-only test trace and finish setup.
Blogs
Merlin: Context-aware Local Detection of Unsafe Agent Tool Calls
Merlin evaluates agent tool calls locally using the user request, tool history, and tool schemas to detect unsafe actions before execution.
2026-09-21 · 8 min read
Guides to runtime agent security
Pricing
Free: $0; 5,000 transactions a month. Team: $500/month; 100,000 transactions, then $1 per 1,000. Enterprise: Custom; Unlimited transactions.
Enterprise pricing is custom. Contact sales for a rollout tailored to your organization.