Skip to content
← Back to blog
Announcement

InnerWarden Joined Claude Startups

October 6, 2026·3 min read

What happened

On 6 October 2026 InnerWarden became a member of Claude Startups, Anthropic's program for early-stage founders. Members get access to offers, tools and events that help them run their business using Claude.

What the program gives

For a company of our size, membership brings:

  • a year of Claude Team for the people who build InnerWarden;
  • a one-time grant of API credits;
  • the Claude Startup Stack, a set of offers on AI-native tools;
  • group sessions with Anthropic's Applied AI team every other week;
  • invitations to Claude Startup events, hackathons and meetups.

Why it matters to us

InnerWarden is runtime security for AI agents, and Claude Code is one of the agents it guards: the free Community guard screens the shell commands Claude Code tries to run, before they run, and on Linux, Active Defence, once armed, refuses a program that is not on the agent's allow list.

Building that well means working with agents every day, the way our users do. The program gives us more room to do that, and the Applied AI sessions are where we plan to bring the hard questions, such as how to evaluate the local decision model we ship.

What it is not

Membership of Claude Startups is not a partnership, a certification or an endorsement. Anthropic has not reviewed or approved InnerWarden. Anthropic, Claude and Claude Code are trademarks of Anthropic, PBC.

Related reading

Keep following the attack path

AI Security

Claude Code Hooks: Security Guide and Examples

Claude Code hooks for security: how a PreToolUse hook blocks a tool call, two working examples, what a hook cannot stop, and what the Linux kernel adds.

9 min readRead
AI Security

How Shell Rewrites Beat Command Blocklists

Attackers defeat text command filters by rewriting a command into a form that means the same thing but does not match the string. The fix is two layers: normalize before matching, and read the real argv in-kernel at execve.

8 min readRead
AI Security

Agentic AI Security Risks, Honestly Mapped

Most tools claim OWASP Agentic Top 10 coverage from the prompt layer. InnerWarden maps to all ten at the action layer with a reason chain, redaction, and a circuit breaker, and is honest about what is proven versus designed.

7 min readRead
AI Security

Indirect Prompt Injection: From Poisoned Input to Syscall

A poisoned input rewrites the agent's intent, and that intent lands on the host as execve, openat, or connect. Prompt-layer defenses live inside the context that just got poisoned. The damage becomes real at the syscall.

7 min readRead
AI Security

Why the Kernel's No Beats the Model's No

A model's refusal is a prediction an attacker can steer. The operating system's refusal is not a decision at all. Where enforcement has to live when agents touch real machines.

6 min readRead
Product

Protect AI Agents at Runtime, Down to the Kernel

Screen every command and tool call an agent makes on any OS; on supported Linux hosts, the Execution Gate refuses an unlisted program at exec.

Open