2026-10-06
Day 6: Pool Party
Lane: Binary reverse engineering. Modes: Single Solve, Time Trial, Mastery. Result: all CORRECT.
The challenge
- ID:
binary-reverse-engineering-day-06-java-bytecode-loader - Prompt: "Every Java class file carries its own phone book: the constant pool. Names, methods, invocations — it's all in there. Dive in and take attendance."
What we were given
- One
evidence.zip(1,919,090 bytes, SHA-256f3f8f329eb707d24bf36afc2e3fadf7f1bdc3f12cd8c7d5a7dbaaf435da7f8b6). AES encrypted, passwordinfected. - The earlier files again:
d01(C),d02(C++),d03(.NET),d04(Go),d05(Rust). - New today:
d06--variant-000.jar(3,099 bytes), a small Java archive. - It holds
Day06Variant000.class,Day06Variant000$Result.classand a manifest withMain-Class: Day06Variant000.
What was asked
- 48 questions in Single Solve and Time Trial. Mastery had two deliveries, so 96.
- Question types:
- Hashes: SHA-256 of each member.
- d01 to d05: the same header, structure and config questions as before.
- d06 (new): the archive format (JAR / Java archive), the config fields
campaign,domain,mutex,url, the CRC32, and theHCFGversion byte.
How it was solved, start to finish
- Ran the driver
run_tm.py: start the attempt, download the zip, runsolve_tm.py, submit. - Routine fields for d01–d05 were answered as before.
- d06 (the new part): the config is not stored as raw bytes. It is a Base64 string inside the class file's constant pool. The question spelled out the method:
- Open the JAR and the
Main-Classclass file. - Walk the constant pool. Take each UTF-8 constant that is at least 80 Base64 characters long.
- Base64-decode it. Accept the first result that is a CRC32-valid
HCFGrecord. - Decode the record as before: XOR the payload with the key byte, check the CRC32, parse the JSON.
The solver opens each JAR member and tests every Base64 run of 40+ characters, which finds the same string. Re-checked on the saved sample:
[jar:Day06Variant000.class base64 run at file offset 160, run length 508]
HCFG version=1 xor_key=0x71 len=368 crc32=0x121bc931
JSON: {"campaign":"BR06-000","domain":"node-000.day06.example.invalid","ipv4":"198.51.100.120","url":"https://node-000.day06.example.invalid/api/BR06-000","mutex":"Global\\HuntressBR06_000", ...}
Answers: format Java archive, CRC32 0x121bc931, HCFG version 1, campaign BR06-000.
- Submitted all 48 answers (
answers.json). Result:CORRECT. - Time Trial: a fresh variant, same steps. Result:
CORRECT. - Mastery: two deliveries (96 answers), same steps. Result:
CORRECT.
Results
| Mode | Result | Flag |
|---|---|---|
| Single Solve | CORRECT | f7ab2726e3bc43378a8909c5f205a4aa |
| Time Trial | CORRECT | e9eaac6fdc6e734018c35bc3b9e4a420 |
| Mastery | CORRECT | 3430a0faa511c2482e0db977e3aab769 |
Each mode folder has the raw start-response.json, answers.json, submit-response.json and final-status-response.json.
Notes on the saved files:
mastery/answers.jsonwas saved empty.mastery/final-status-response.jsonshows all 96 answers accepted.mastery/start-response.jsonis from a later restart that the Game Master refused (rotation_exhausted).
Reproduce
Offline, from this folder (no Game Master needed):
mkdir -p /tmp/d06/deliveries && cp single-solve/evidence.zip /tmp/d06/deliveries/delivery-1.zip
python3 -I ~/.config/huntress/solvers/solve_tm.py single-solve/start-response.json /tmp/d06/deliveries /tmp/d06/answers.json
Checked on 2026-11-01: this gives the same 48 answers as answers.json. It only parses the samples and never runs them.
Live driver:
python3 -I ~/.config/huntress/solvers/run_tm.py binary-reverse-engineering-day-06-java-bytecode-loader single_solve /tmp/d06
This needs a Game Master token in ~/.config/huntress/gm-token.
Day 6: Portable Deception
Lane: Metadata forensics. Modes: Single Solve, Time Trial, Mastery. Result: all CORRECT.
The challenge
- ID:
metadata-forensics-day-06-portable-deception - Prompt: "A document bundle mixes real PDFs, renamed impostors, and multiple declared versions. Determine what the files promise, what they contain, and which promise is false."
What we were given
- Single Solve:
(evidence.zipmetadata-day-06-evidence.zip, 788 bytes). It holds(1,552 bytes).document-001.pdf - Time Trial:
metadata-day-06-evidence.zip(788 bytes, SHA-2563dbdcc5e0729767e10b3e7fc33fb0d2d59561b6e74c0ce6b159c5973840610fa) withdocument-002.pdf. - Mastery:
metadata-forensics-day-06-portable-deception-mastery.zip(1,871 bytes, SHA-256275284f4a84beb0d6b9a3ad4c999071838e8cfaa244e8124ae28e035b8b7804b). It holdsmetadata-day-06-evidence-01.zip(document-003.pdf) andmetadata-day-06-evidence-02.zip(document-004.pdf). - New today: PDF files.
What was asked
- One field type:
pdf.header_versionfor a named PDF. - The answer is the version number from the header, like
1.4. - Single Solve and Time Trial: 1 question. Mastery: 2 questions (
set1.,set2.). - Timers: Time Trial 50 seconds, Mastery 80 seconds.
How it was solved, start to finish
The method: a PDF declares its version in its first bytes as %PDF-x.y. Read that, not the file name.
- Single Solve. Downloaded the zip, extracted
document-001.pdfand looked at the first bytes:
%PDF-1.4
Submitted 1.4 as d06-pdf-001-pdf-header-version. CORRECT.
- Added a
pdf_versionsolver to therun_mode.pydriver:
def pdf_version(b):
m = re.match(rb'%PDF-(\d+\.\d+)', b[:16])
return m.group(1).decode() if m else None
- Time Trial. The driver answered
document-002.pdf=1.5.CORRECT. - Mastery. The driver found each named PDF inside its inner zip. Answers:
document-003.pdf=1.6,document-004.pdf=1.7.CORRECT.
Results
| Mode | Result | Flag |
|---|---|---|
| Single Solve | CORRECT | 426b37cd5a9256c51d79870949714013 |
| Time Trial | CORRECT | 9e82f182fc38871a49639f3b515c1e1b |
| Mastery | CORRECT | cc014843b1d13ceff35684478b1530b3 |
Each mode folder has answers.json, start-response.json, submit-response.json and final-status-response.json. The Single Solve folder also has the zip and the PDF; its saved submit says already_solved because it is a re-run, and the final status shows CORRECT. No screenshots were captured.
Reproduce
python3 -I ~/.config/huntress/solvers/run_mode.py metadata-forensics-day-06-portable-deception single_solve pdf_version /tmp/d06
This needs a Game Master token in ~/.config/huntress/gm-token. For time_trial and mastery the driver calls op=retry first, which spends one of the unique attempts. It only reads the file header.
Day 6: Provider Paths
Lane: Mobile (Android). Modes: Single Solve, Time Trial, Mastery. Result: all CORRECT.
The challenge
- ID:
mobile-android-day-06-provider-paths - Prompt: "Abuse an overbroad provider path and correlate its records with prior intent evidence."
What we were given
- One Android APK per case. Inside:
AndroidManifest.xml,classes.dex,assets/huntress_puzzle.json, andMETA-INF/signing files. - Single Solve:
provider_paths.apk, caseMOB-06-E69BAAF4FE(12,704 bytes, SHA-2562bbfcd435f49c58d9a9aa54f9a3d2d04b7601114d16b0a943a795d2675645a12). Its puzzle is saved assingle-solve/huntress_puzzle.json. - Time Trial: one APK. Mastery: three APKs. These APKs were not saved in the repo.
What was asked
- One question per case: return the exact backend answer code for that case.
- Answer format:
ANS-plus exactly 16 uppercase hex characters. - Answer IDs:
answerfor Single Solve and Time Trial.set1.answertoset3.answerfor Mastery. - Time limits: Time Trial 50 seconds, Mastery 110 seconds.
How it was solved, start to finish
The app was never installed or run.
- Started the Single Solve and downloaded the APK.
- Read the puzzle asset. New codec today:
hex.
{"schema":"huntress.mobile.puzzle.v1","day":6,"codec":"hex","data":"414e532d34364132344243444145354336453142"}
- Decoded it. Each pair of hex digits is one ASCII character:
import json
p = json.load(open('single-solve/huntress_puzzle.json'))
print(bytes.fromhex(p['data']).decode())
Output: ANS-46A24BCDAE5C6E1B.
- Submitted it under
answer. The Game Master repliedCORRECT. - Taught the driver. Added the
hexcodec tomobile_puzzle()inrun_mode.py. - Time Trial and Mastery: ran
run_mode.pyon both. Both came backCORRECT.
Results
| Mode | Result | Flag |
|---|---|---|
| Single Solve | CORRECT | a7fa92197d6669d1175432fbbe6cf87f |
| Time Trial | CORRECT | 064fb96bb819f5a6b6555a5f0349c276 |
| Mastery | CORRECT | e76211b02a3b3b996bcdfed5b63b367f |
Each mode folder has answers.json, start-response.json, submit-response.json and final-status-response.json. The Single Solve submit-response.json says already_solved because it is a resubmit. Its final-status-response.json holds the CORRECT result and flag. No screenshots were taken.
Reproduce
Offline: run the Python snippet above from this folder.
A fresh run with the driver:
python3 -I ~/.config/huntress/solvers/run_mode.py mobile-android-day-06-provider-paths single_solve mobile_puzzle /tmp/d06
This needs a Game Master token in ~/.config/huntress/gm-token. It only unzips the APK and reads the asset. It never installs or runs the app.
Day 6: Shell We Begin
Lane: Script analysis. Modes: Single Solve, Time Trial, Mastery. Result: all CORRECT.
The challenge
- ID:
script-analysis-day-06-bash-behavior-graphs - Prompt: "Six Bash behavior graphs, none of which will ever run. Read them like a native: map the adapters, count the stages, tag the ATT&CK techniques, and pull the exact indicators hiding in the final bytes."
What we were given
- One
evidence.zip(56,162 bytes, SHA-25641694eb79f07dd9fceba8e5b51955ccb38b9db91708d7c97d5ae57e7bf473387). Password:infected. - The Day 5 set plus 6 new Bash files:
host-context.sh,memory-stage.sh,embedded-write.sh,remote-write.sh,local-dispatch.sh,staged-chain.sh. - Each Bash file uses the same pattern as VBScript (Day 4) and Python (Day 5), with single-quoted strings. Real lines from
embedded-write.sh:
readonly Material_7dd3e2e1='72 29 78 74 78 67 76 26 26 30 86 30 28 26 15 8 ...'
readonly Codec_7dd3e2e1='decimal_space|reverse_then_xor|7b'
What was asked
- 399 questions: the 357 from Day 5 plus 42 Bash questions (7 per file).
- Bash questions per file:
- Role and ATT&CK ID.
- Steps: the ordered behavior graph as tokens joined by
>. - Stage count (an integer), adapter and campaign ID.
- One indicator: URL, file path, cron path or host ID.
How it was solved, start to finish
- Started the untimed Single Solve and downloaded the archive.
- Reused
solve_expansion.pyfor the 357 older questions. - Wrote a Bash decoder (
bashdec.py):
- It joins the quoted pieces of
Material_(Bash glues'a''b'together, so the solver does too). - It reads
Codec_and parses the material by its representation. New today:percent_hex(%41%42...). - Then it runs the same transform search as Day 4. The result must match
v=<n>|<key>=.
Example: embedded-write.sh decodes to:
v=1|steps=decode_embedded,write_file|adapter=printf_decode_redirect|stage_count=1|campaign_id=campaign-eaa78515f3|file_path=/tmp/huntress-cache/.stage-eaa78515f3
- Mapped answers. Role comes from the steps, as on Days 3–5. Bash's Command and Scripting Interpreter ID is
T1059.004. The questions map to files through theirartifactfield. - Checked with the partial-submit oracle (see Day 3). Only
bash:steps(6 answers) was rejected. The value format was wrong: the question wants the steps joined with>and no spaces. After joining them with>, the Single Solve wasCORRECT.
| File | Steps (as submitted) | Role | ATT&CK |
|---|---|---|---|
host-context.sh | decode_configuration>return_configuration | benign_utility | none |
memory-stage.sh | acquire_remote>dispatch_pipe | stager | T1105 |
embedded-write.sh | decode_embedded>write_file | dropper | T1140 |
remote-write.sh | acquire_remote>write_file | downloader | T1105 |
local-dispatch.sh | locate_local>dispatch_process | launcher | T1059.004 |
staged-chain.sh | profile_host>acquire_remote>write_file>change_mode>dispatch_process | staged_chain | T1059.004 |
A few more saved answers:
{"bash-dropper-adapter": "printf_decode_redirect", "bash-dropper-stage-count": "1",
"bash-dropper-file-path": "/tmp/huntress-cache/.stage-eaa78515f3",
"bash-chain-cron-path": "/etc/cron.d/update-9ea2805736"}
- Time Trial (50-second deadline) and Mastery (798 questions in two sets, 80-second deadline) were run with
run_expansion.py. Both wereCORRECTon the first run.
Results
| Mode | Result | Flag |
|---|---|---|
| Single Solve | CORRECT | eee6b42c0ebd4c66594522fcc5470d15 |
| Time Trial | CORRECT | cd4144095535d690b2eb378ecdfa1bcb |
| Mastery | CORRECT | 6f58d5d6e5b0632917a6ee5ae0d571ff |
Each mode folder has answers.json and the Game Master JSON. The Single Solve folder also keeps evidence.zip. No screenshots were taken.
Reproduce
Offline, from the repo root:
mkdir -p /tmp/d06/ex
7z x -pinfected -o/tmp/d06/ex repos/day-06-script-analysis-shell-we-begin/single-solve/evidence.zip
python3 -I tools/solve_expansion.py /tmp/d06/ex \
repos/day-06-script-analysis-shell-we-begin/single-solve/start-response.json /tmp/d06/answers.json
This rebuilds all 399 saved answers (re-checked on 2026-11-01). It only parses the files.
Live: python3 -I ~/.config/huntress/solvers/run_expansion.py script-analysis-day-06-bash-behavior-graphs <mode> <workdir>. This needs a Game Master token. The solvers/ copy runs JScript fixtures in a Node vm sandbox; the tools/ copy does not.
Day 6: Eight Is Enough
Lane: Warmups. Modes: Single Solve, Time Trial, Mastery. Result: all CORRECT.
The challenge
- ID:
warmups-day-06 - Prompt: "Some traditions come in groups of three. Eight values is enough to say it all."
- New encoding: octal ASCII
What we were given
- One blob file per mode (Single Solve and Time Trial), or a zip of multiple blobs (Mastery).
- Each blob has numbered lines (
NN<TAB>fragment). Each line is independently encoded. - The decoded text ends with a slug like
word-word-word-NNto submit.
How it was solved
- Decode the blob. Space-separated octal numbers. Each is an ASCII code; convert from base 8.
- Mastery blobs use multiple encodings from Days 1-6. The solver (
warmup_decode.py) tries every known decoder on each line and picks the one producing printable text. - Concatenate all decoded lines. The text contains a message ending in the answer slug.
- Submit the slug to the Game Master.
Results
- Single Solve:
CORRECT. slug:hearty-raven-clover-39. Flag:9466db6842d694d1ee1364d55244a621. - Time Trial:
CORRECT. slug:tidy-castle-clover-76. Flag:e94ecb8b31ebe2689041afff7fcfcba1. - Mastery:
CORRECT. 28 slugs. Flag:23b6f2cbcce7a24cd810fb444e46cc66.