native library secrets: a codec named native
huntress ctf 2026, day 15 · mobile android · all three modes
tools used: android apk toolkit and timed-attempt pipeline.
the prompt: "recover split configuration from dex and a challenge-owned native library." the apk carried a native library for the first time, and its puzzle named a codec no earlier day used.
the key insight: a new codec name is not a new transform. native named no transform at all: the puzzle's data was the code itself, stored verbatim, and the same string sits in the native library. an unknown codec name leaves the solver no step to run, so the first reading it tests is the data as given, and that reading already had the answer's shape.
challenge shape
single 1 field one apk
time trial 1 field another apk
mastery 5 fields five apks
the field is answer (set1.answer to set5.answer in mastery), with the usual syntax: "ans- followed by exactly 16 uppercase hexadecimal characters; no spaces or explanatory text".
artifact anatomy
AndroidManifest.xml
classes.dex
assets/huntress_puzzle.json {"schema": "huntress.mobile.puzzle.v1", "day": 15, "codec": "native", "data": "ANS-E28C9614AEF73B8F"}
lib/x86_64/libarena.so the challenge's own native library
the library is an unstripped x86-64 elf shared object; its read-only data holds one ANS- string, the same one the json stores.
the mastery's five apks replayed earlier days' puzzles, and each json still carried its source day in day: set 1 was day 3's base64, set 2 day 7's keyed caesar, sets 3 and 4 the base64 of days 9 and 13, and only set 5 was a day 15 native apk with the library.
extraction method
the puzzle names its codec; a known name selects its transform and key source (the json, the dex strings, the native library's strings). an unknown name such as native makes the codec parser raise "unsupported puzzle codec", the step list is empty, and the chain yields the data unchanged; had that not matched, the keyed search and the generic peel would have run next. by hand, either source works:
unzip -p native_library_secrets.apk assets/huntress_puzzle.json
unzip -p native_library_secrets.apk lib/x86_64/libarena.so | strings | grep -E '^ANS-[0-9A-F]{16}$'
validation and answer formatting
a decoded value is accepted only when it matches the question's pattern syntax; anything else is a crash, never a guess.
automation across modes
the same decoder answered the single, the time trial and all five mastery sets; each set's codec comes from its own puzzle.
lessons
- name dispatch with a search fallback absorbs a new codec name with no code change, as long as the result can be checked against the answer's shape.
solve times
public scoreboard · day 15 finish: top 10
- 0:03.2singlefirst blood, +1.3 s
- 0:04.9time trialfirst blood, +0.88 s
- 0:06.6masteryfirst blood, +0.83 s
- #4team day finishof 163 teams that cleared it, in 2:31.2
- #2single orderof the day's 5, by our flag time
from the public scoreboard: 09:00 et release to our flag; the table below counts from our attempt's start.
from when we started working on the mode to its flag.
| mode | work to flag | attempts | wrong submits |
|---|---|---|---|
| single | 1.7s | 1 | 0 |
| time trial | 1.7s | 1 | 0 |
| mastery | 1.7s | 1 | 0 |
the clock
server timer of the winning attempt; used is attempt start to flag, left is what remained on the clock.
| mode | time available | time used | time left |
|---|---|---|---|
| time trial | 50.0s | 0.6s | 49.4s |
| mastery | 170.0s | 0.8s | 169.2s |
where we got stuck
- nowhere: every mode passed on its first attempt.
flags: single, time trial, and mastery [redacted].
{generated by a clanker}