SKELETONKEY

leviathan/SKELETONKEY

Fork 0

Commit Graph

Author	SHA1	Message	Date
leviathan	3b287f84f0	copy_fail_family: skip DIRTYFAIL typed prompt under --i-know The vendored DIRTYFAIL exploits call typed_confirm("DIRTYFAIL"), which reads stdin interactively. SKELETONKEY already gates --exploit/--auto behind --i-know, so the prompt is redundant and deadlocks non-interactive runs like `skeletonkey --auto --i-know`. Add a dirtyfail_assume_yes flag, forwarded from skeletonkey_ctx.authorized by the bridge layer's apply_ctx(). When set, typed_confirm() auto-satisfies its gate and logs that it did so. The YES_BREAK_SSH self-lockout guard is exempt — it protects the operator's own access rather than gating authorization, so it still requires an interactive answer. Standalone DIRTYFAIL builds are unchanged: the flag defaults false.	2026-05-22 16:49:15 -04:00
leviathan	52e8c99022	Phase 1: module interface + registry + top-level dispatcher - core/module.h: struct iamroot_module + iamroot_result_t - core/registry.{h,c}: flat-array module registry with find-by-name - modules/copy_fail_family/iamroot_modules.{h,c}: bridge layer exposing 5 modules (copy_fail, copy_fail_gcm, dirty_frag_esp, dirty_frag_esp6, dirty_frag_rxrpc) wired to the absorbed DIRTYFAIL detect/exploit functions; df_result_t/iamroot_result_t share numeric values intentionally for zero-cost translation - iamroot.c: top-level CLI dispatcher with --scan / --list / --exploit / --mitigate / --cleanup, JSON output, --i-know gate - Restored modules/copy_fail_family/src/ structure (DIRTYFAIL Makefile expects it; the initial flat copy broke that contract) - Top-level Makefile builds one binary; filters out DIRTYFAIL's original dirtyfail.c main so it doesn't conflict with iamroot.c Verified end-to-end on kctf-mgr (Linux): clean compile, 5 modules register, --scan --json output ingest-ready, exit codes propagate.	2026-05-16 19:32:11 -04:00

Author

SHA1

Message

Date

leviathan

3b287f84f0

copy_fail_family: skip DIRTYFAIL typed prompt under --i-know

The vendored DIRTYFAIL exploits call typed_confirm("DIRTYFAIL"), which
reads stdin interactively. SKELETONKEY already gates --exploit/--auto
behind --i-know, so the prompt is redundant and deadlocks non-interactive
runs like `skeletonkey --auto --i-know`.

Add a dirtyfail_assume_yes flag, forwarded from skeletonkey_ctx.authorized
by the bridge layer's apply_ctx(). When set, typed_confirm() auto-satisfies
its gate and logs that it did so.

The YES_BREAK_SSH self-lockout guard is exempt — it protects the
operator's own access rather than gating authorization, so it still
requires an interactive answer.

Standalone DIRTYFAIL builds are unchanged: the flag defaults false.

2026-05-22 16:49:15 -04:00

leviathan

52e8c99022

Phase 1: module interface + registry + top-level dispatcher

- core/module.h: struct iamroot_module + iamroot_result_t
- core/registry.{h,c}: flat-array module registry with find-by-name
- modules/copy_fail_family/iamroot_modules.{h,c}: bridge layer
  exposing 5 modules (copy_fail, copy_fail_gcm, dirty_frag_esp,
  dirty_frag_esp6, dirty_frag_rxrpc) wired to the absorbed DIRTYFAIL
  detect/exploit functions; df_result_t/iamroot_result_t share numeric
  values intentionally for zero-cost translation
- iamroot.c: top-level CLI dispatcher with --scan / --list / --exploit /
  --mitigate / --cleanup, JSON output, --i-know gate
- Restored modules/copy_fail_family/src/ structure (DIRTYFAIL Makefile
  expects it; the initial flat copy broke that contract)
- Top-level Makefile builds one binary; filters out DIRTYFAIL's
  original dirtyfail.c main so it doesn't conflict with iamroot.c

Verified end-to-end on kctf-mgr (Linux): clean compile, 5 modules
register, --scan --json output ingest-ready, exit codes propagate.

2026-05-16 19:32:11 -04:00

2 Commits