Skip to content

Security: dotgibson/dotfiles-Offense

SECURITY.md

Security Policy

dotfiles-Offense is a public repository that ships offensive tooling configuration. It contains no exploit code, no payloads, and no engagement data — but it does configure tools that reach the network, and it installs software on whatever machine runs bootstrap.sh. That makes it worth a disclosure path of its own.

dotfiles-core's SECURITY.md explicitly puts this layer out of its scope and directs layer-specific reports here.

Reporting a vulnerability

Please use GitHub's private vulnerability reporting — it opens a private advisory thread visible only to the maintainers.

Do not open a public issue for anything in the "in scope" list below.

If private reporting is unavailable to you, email the maintainer address in README.md and put SECURITY in the subject.

Expect an acknowledgement within a week. This is a personal project maintained in spare time, so there is no formal SLA and no bounty.

In scope

Things that would genuinely compromise a machine running these dotfiles:

  • Supply chain in bootstrap.sh — an unverified download, a writable path used before verification, a step that can be induced to run attacker-controlled code. --install is the only path that fetches anything: apt on Kali, and pipx/go elsewhere, both of which verify against PyPI hashes and the Go checksum database. A step that bypasses either is a valid report.
  • Privilege escalation through the install path — anything that widens what sudo is used for, a sudo invocation on an attacker-influenced path, or a world-writable artifact left behind.
  • Symlink handlingbootstrap.sh and core/lib/bootstrap-lib.sh create links in $HOME. A path traversal, an unintended clobber, or a symlink cycle is in scope.
  • Engagement-data leakage — anything that could route client data into the repository. The helpers in offensive/offensive.zsh refuse to write inside a git work tree without $ENGAGEMENT, and .gitignore is the backstop; a bypass of either is a valid report and is treated as high severity.
  • Secrets in history or in the tree — despite the gitleaks gate in .github/workflows/checks.yml.
  • CI/workflow issues — script injection into a workflow, an over-permissioned GITHUB_TOKEN, a mutable action reference that should be pinned.

Out of scope

  • The offensive tools themselves. nmap, nxc/NetExec, impacket, sliver, Metasploit and friends are third-party software. Report vulnerabilities in them to their own projects; this repo only lists and configures them.
  • "This repo enables attacks." That is the stated purpose: it is tooling for authorized engagements under written rules of engagement. See OFFENSIVE-METHODOLOGY.md.
  • The vendored subtrees. core/ is a copy of dotfiles-core and offensive/companion/ is a copy of htpx. Report against those repositories — a fix here would be overwritten on the next sync. If you are unsure which layer owns the bug, report it here and it will be routed.
  • Findings from an automated scanner with no demonstrated impact on a machine running these dotfiles.

Supported versions

Only the tip of the default branch. This is a rolling configuration repository: tags exist to record what was vendored at a point in time, not to designate a maintained release line. Fixes land on main.

There aren't any published security advisories