Skip to content

Generate VM's with kernel tracing, code sandboxing and security profiles for long running agents.

License

Notifications You must be signed in to change notification settings

Ghostlock-AI/capsule

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

358 Commits
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

Capsule


TL;DR

Capsule watches agent behavior from the kernel (eBPF/LSM), enriches events into human-readable timelines, and lays the groundwork for dynamic, policy-driven security backed by small ML models. It’s pre-alpha, Linux aarch64 only right now, written in Rust.

License

This project is licensed under the MIT License.


Why we are building Capsule

Capsule is a permanently open-source kernel-level tracing and sandboxing project.
It will never be closed, dual-licensed, or converted into a proprietary core.

Here at Ghostlock, we are building Capsule because application-level observability and enforcement no longer hold up in a world of autonomous agents. User-mode hooks are easy to bypass, and existing tools cannot reliably explain what code actually did once it runs.

As software shifts toward autonomous agents—systems that write code, spawn processes, and make decisions with minimal supervision—the human role moves from approving outputs to understanding and constraining behavior.

Kernel-level tracing provides a stronger foundation. By observing execution below the application layer, it becomes possible to produce durable, verifiable records of behavior that software cannot evade. Today, however, these tools remain fragmented, hard to use, and accessible only to specialists.

Capsule exists to make kernel-level tracing practical:

•	simple enough to experiment with,

•	explicit enough to reason about, and

•	open enough to trust long-term.

If you work on kernels, runtimes, security, or systems tooling—and care about making autonomous software observable and accountable—we welcome your contributions!


What Capsule observes

Area In plain terms
Process execution When programs start, fork, or become backgroud processes
Network All network communication—who talks to whom.
File I/O Read/write/create/delete/move files and folders.
Credentials Changes to identity (UID/GID/capabilities).
Memory / code Risky mappings (e.g., W+X), code loading.
IPC orchestration Local process-to-process comms (pipes, UNIX sockets, etc.).
Device access Access to /dev/* (KVM, tun/tap, GPU, disks, USB/TTY).
System configuration Mounts, chroot/pivot_root, persistence paths.
Containers & cgroups Enter/leave namespaces; resource limits and cgroup changes.
Signals Software interrupts (SIGKILL, SIGTERM, etc.).

Architecture

  • Kernel Probes: eBPF kprobes/tracepoints/LSM hooks (Linux) capture syscall-level and semantic events.
  • Userspace Daemon: stream ingestion, async enrichment of syscalls for better readability.
  • Policy/ML Layer: deterministic rules + sequence/graph model that categorizes prompt, syscall sequence, and resource utilization combinations as risky or harmless.

About

Generate VM's with kernel tracing, code sandboxing and security profiles for long running agents.

Topics

Resources

License

Stars

Watchers

Forks

Packages

No packages published

Contributors 3

  •  
  •  
  •  

Languages