Sumar
In captura de retea exista doua fluxuri TCP mari:
- un flux de tip zgomot pe portul
80(noise.local), cu multe POST-uri de 1 MB; - un flux real pe portul
5000, care contine un upload HTTP multipart catreoctoprint.local.
Din al doilea flux se recupereaza fisierul chimera_test.gcode (aprox 47 MB). Mesajul nu este in clar in
text, ci este ascuns geometric in traseele de printare din straturile din mijloc (Z=1.0 pana la Z=4.8).
Dupa ce traseele sunt convertite in grila 2D si se vizualizeaza void-ul (inversul materialului extrudat),
apare flag-ul.
Dovada rezolvarii
Cand deschidem fisierul .pcap in Wireshark putem vedea in statistici ca sunt doar pachete TCP:

Acuma folosind tshark:

Observatii:
- Captura este TCP-only.
- Sunt doua conversatii principale:
192.168.1.99:33333 -> 192.168.1.10:80(aprox 27 MB)192.168.1.50:44444 -> 192.168.1.10:5000(aprox 49 MB)
Concluzie: fluxul de pe 5000 este mai mare si mai suspect. Pe portul 80 apar request-uri de forma
POST /noise/data_0 … POST /noise/data_24, fiecare cu Content-Length: 1048576. Acest model
repetitiv indica trafic de umplutura (decoy).
Inspectia primului pachet cu date din fluxul pe 5000 arata un upload direct de G-code catre un server
de tip OctoPrint:
POST /api/files/local HTTP/1.1Host: octoprint.localContent-Type: multipart/form-dataContent-Length: 47892095filename="chimera_test.gcode"
Extragem fluxul cu tcpflow:

Acuma, ca sa putem vizualiza layerele am facut un script de python care genereaza un fisier .png:
import math
import importlib
import re
import sys
# Fixed challenge defaults.
LAYER = 1.0
EXTENT = 650.0
CELL = 5.0
SCALE = 8
X_RE = re.compile(r"\bX(-?\d+(?:\.\d+)?)")
Y_RE = re.compile(r"\bY(-?\d+(?:\.\d+)?)")
def main() -> None:
try:
image_mod = importlib.import_module("PIL.Image")
except ModuleNotFoundError:
print("[!] Pillow is required. Install with: pip install pillow")
sys.exit(1)
if len(sys.argv) < 2:
print("Usage: python3 generate_chimera_pngs.py <gcode_or_stream_file> [output_png]")
sys.exit(1)
input_file = sys.argv[1]
output_png = sys.argv[2] if len(sys.argv) > 2 else "layer1_material.png"
size = int(round(EXTENT / CELL))
grid = [[0] * size for _ in range(size)] # 1 = material
z = None
cur_x = None
cur_y = None
with open(input_file, "r", errors="ignore") as f:
for line in f:
line = line.strip()
if line.startswith("G0 Z"):
z = float(line.split("Z", 1)[1])
continue
if z != LAYER or not line.startswith(("G0 ", "G1 ")):
continue
x_match = X_RE.search(line)
y_match = Y_RE.search(line)
x = float(x_match.group(1)) if x_match else cur_x
y = float(y_match.group(1)) if y_match else cur_y
if (
line.startswith("G1 ")
and cur_x is not None
and x is not None
and y is not None
and x != cur_x
):
row = int(y // CELL)
if 0 <= row < size:
x0 = min(cur_x, x)
x1 = max(cur_x, x)
c0 = max(0, int(x0 // CELL))
c1 = min(size - 1, int(math.ceil(x1 / CELL) - 1))
for col in range(c0, c1 + 1):
grid[row][col] = 1
cur_x, cur_y = x, y
img = image_mod.new("L", (size, size), 255)
for row in range(size):
for col in range(size):
if grid[row][col]:
img.putpixel((col, size - 1 - row), 0)
img = img.resize((size * SCALE, size * SCALE), image_mod.NEAREST)
img.save(output_png)
print(f"[+] Wrote {output_png}")
if __name__ == "__main__":
main()
Rulam generatorul peste stream-ul reasamblat:

Void-ul dintre trasee deseneaza flag-ul:
