Skip to content
KitploitKITPLOIT
ToolsExploitsBlog
Log in
Submit
ToolsExploitsBlog
Submit

Hacking, PenTest, and Cybersecurity Tools for Your Security Arsenal!

Kitploit is a directory of hacking, cybersecurity, and pentesting tools. Discover the latest project updates to find vulnerabilities, analyze systems, automate testing, and strengthen your security.

FeedsContactPrivacy© 2026 Kitploit

Tool Directory

Categories

View all categories
Loading categories
Fragnesia-go — A Go implementation of fragnesia (CVE-2026-46300) | Kitploit
Tools/GitHubGitHub/nonameuserosint-hue/fragnesia-go
Privilege EscalationExploitationBinary Exploitation
GitHubnonameuserosint-hue/fragnesia-go

Fragnesia-go

A Go implementation of fragnesia (CVE-2026-46300)

View Repository
256 months agoNot yet reviewed
Share

fragnesia-go

One SA. One byte at a time. No race.

Go port of fragnesia (CVE-2026-46300). Member of the Dirty Frag vulnerability class - separate bug, same surface, own patch.

How it works (tl;dr)

The bug is in the Linux XFRM ESP-in-TCP subsystem. When TCP_ULP espintcp is installed on a socket after data has already been splice()d from a file into the TCP receive queue, the kernel treats the queued data as an ESP ciphertext record and decrypts it in-place. The source of that data is a page-cache page. The decryption XORs the AES-GCM keystream directly into it.

The result is a controlled single-byte write into the page cache of any readable file.

Keystream table

AES-GCM (RFC 4106) counter block at position 2 is [salt(4) || IV(8) || 0x00000002]. Encrypting it under the session key gives a 16-byte keystream block. Byte 0 of that block is what gets XORed into the target byte. By varying the lower 32 bits of the IV nonce, all 256 possible keystream byte values are reachable within the first 65536 nonces. The exploit builds this lookup table once at startup via AF_ALG AES-ECB.

Byte-flip loop

For each byte of the payload:

  1. Read the current value from the file (page cache).
  2. Compute needed_keystream = current XOR desired.
  3. Look up the nonce for that keystream byte.
  4. Set the IV, fire a TCP trigger pair (sender splices the file page into the socket, receiver delays TCP_ULP install until the data is queued).
  5. Result: current XOR keystream = current XOR (current XOR desired) = desired.

One SA. 192 triggers maximum. Deterministic.

Usage

go build -o fragnesia-go .
./fragnesia-go

On success it drops into a root shell via a fresh PTY. The on-disk binary is untouched - page cache only.

Cleanup

echo 1 | tee /proc/sys/vm/drop_caches

Requirements

  • Linux - unpatched kernel (see below)
  • No external tools - pure Go, single static binary
  • Kernel module: esp6 (transport mode, IPv6 loopback)

Affected kernels

All kernels before the patch: https://lists.openwall.net/netdev/2026/05/13/79

Same range as dirtyfrag.

Mitigation

Same as dirtyfrag:

rmmod esp4 esp6
printf 'install esp4 /bin/false\ninstall esp6 /bin/false\n' \
  > /etc/modprobe.d/fragnesia.conf

Compared to dirtyfrag-go

dirtyfrag-gofragnesia-go
CVEsCVE-2026-43284 / CVE-2026-43500CVE-2026-46300
TransportESP-in-UDP + RxRPCESP-in-TCP (ULP)
Write granularity4 bytes per trigger1 byte per trigger
XFRM SAs481
CryptoHMAC-SHA256 + CBC-AESAES-128-GCM
IP familyIPv4IPv6

References

  • v12-security/pocs - fragnesia - original C PoC
  • CVE-2026-46300 - Red Hat
  • Kernel patch

Credits

  • William Bowling / V12 team - vulnerability discovery and original PoC
Download Tool

Most Popular

View all →

Discover the most used tools by our community.

  • 1

    OsintHunter

    letox4/osinthunter

    🎯 The ultimate OSINT toolkit: investigation graph, confidence scoring, alias families, map, timeline and reports. 49 modules, 400+ sites, no API keys required.

    916
  • 2

    ghostlock-app

    yukonga/ghostlock-app

    GhostLock One-Tap Execution App (CVE-2026-43499)

    558
  • 3

    redforge

    cybersecurity91/redforge

    Browser-side red team toolkit with 102 offline tools for recon, payload generation, credential attacks, crypto, encoding and engagement reporting.

    370
  • 4

    ARTEX

    baarai/artex

    LLM multi-agent autonomous penetration testing platform with planner/worker agents, asset and exploration graphs, MITM traffic recording, and vulnerability retesting.

    360
  • 5

    looking-glass

    anonghosty/looking-glass

    CSIRT platform that ingests Shadowserver reports into MongoDB and replays them as interactive maps, analysis, findings, advisories and executive dashboards.

    317
  • 6

    CyberStrikeAI

    ed1s0nz/cyberstrikeai

    The system of action for AI-native cybersecurity—where intent becomes governed execution, evidence becomes operational memory, and every operation improves the next.

    289

Explore all tools

Browse our collection of tools

View all tools →