agentleFS
Sign inStart free
NoorQureshiNoorQureshi/SploitAgent/skills/cloud/cloud-container-escape/SKILL.md

cloud-container-escape

Skill26 starsChanged 32 days agoMITView on GitHub

Break out of a container to the host. Load when you have a shell in a container/pod and want the node: privileged container, mounted docker.sock, dangerous capabilities, hostPath, or "escape the container". Signals: /.dockerenv, cgroup shows docker/k8s, CAP_SYS_ADMIN, mounted socket.

2 other projects use a version of this.

Adopted by

Most recently taken up by

The file

We split this skill into the rules it gives an agent

1 rule. About 0 are common, 1 is distinctive, and 0 are very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈1

    Distinctive · estimate

    This author's own choices. Worth a look.

  • 0

    Very specific

    Tied to their project's names, files or tools. Probably skip.

  • Common
  • Distinctive
  • Very specific
  1. Distinctive:

    hostPath / host mounts: write to host filesystem (cron, ssh keys, kubelet config).

    line 37

≈ “Common” is an estimate: at least 1 in 5 of the 350,495 skill files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.

Keep what fits and drop the rest. Your version, ready for your agents.
Files
SKILL.md

Container escape → host

When it applies

You have code execution inside a container and want the underlying host (and from there, other containers / the cluster / cloud creds).

Why it works

Containers are isolation by configuration, not a hard boundary. Privileged mode, extra capabilities, mounted host resources, or an exposed Docker socket give processes inside the container direct control over the host kernel/daemon.

Method

  1. Confirm & enumerate: /.dockerenv, /proc/1/cgroup; run amicontained/deepce to list capabilities, seccomp, and mounts.
  2. Exploit the misconfig:
    • Docker socket mounted (/var/run/docker.sock): docker -H unix://... run -v /:/host ... → mount host FS as root → escape.
    • Privileged / CAP_SYS_ADMIN: mount the host disk, or use the classic release_agent cgroup notify_on_release trick to run a command on the host.
    • hostPath / host mounts: write to host filesystem (cron, ssh keys, kubelet config).
    • hostPID: access host processes; nsenter into PID 1's namespaces.
  3. From the host: grab cloud metadata creds (→ cloud-imds-ssrf), pivot to other containers/nodes (→ cloud-kubernetes).

Gotchas

  • Enumerate capabilities/mounts first — the escape depends entirely on which misconfig exists.
  • Kernel-exploit escapes are last-resort and risky on live targets; prefer the config-based paths.
  • On bug bounty, a proven mount of the host FS is enough — don't disrupt the node.

Verify success

Command execution or file read/write on the host (outside the container namespace) — e.g. host /etc/shadow, other containers, or the kubelet/cloud creds.

References

deepce/amicontained; "Understanding Docker container escapes" (Trail of Bits); HackTricks.

Download now (ZIP)

Does it work?

Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.

No reviews yet

People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.

  • It worked
  • Partly
  • It didn't
What did you use it for? e.g. Reviewing pull requests in a Rails app

Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.

Your trail

Files you open will collect here, so you can find them again.

Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.

Now do this for your own.

Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.

Check your setup, free