- A Little Based
FAST PATH. "Sixty-four is a good base" -> Base64.
- ELF Aware
Static ELF orientation: header fields, a custom section, the symbol table and a constant returned by a function. The binaries were never executed.
- Exported Trust
The decoded AndroidManifest.xml shows that com.huntress.mobile.d01.v073e9a824e9e.ChallengeActivity is exported. Its intent filter handles huntress01://exported-trust/unlock (path prefix /unlock). In the ...
- Power Play
Classification: static PowerShell triage (malware_re / scripting_data). No execution is needed. The challenge text itself says the scripts "will never run". Every specimen uses the same template: a Base64 ...
- Read the Rules
RULE 1 (simplest-first) + RULE 2: a URL challenge -> fetch the page, grep source for the flag before anything heavier. Observation: ctf.huntress.com is a Next.js SPA. The static HTML is just a loader shell ...
- Welcome Robots
AI-native orientation: the answer is obtained by driving the Homebase AI workbench (reachable only via the CourseStack-authenticated web terminal; the 10.x Game Master and bare proxy deep-links are not ...
- You Must Be This Hash to Ride
Hash inventory: identify each file from its bytes, then report the filename, size, MD5, SHA-1 and SHA-256 of the file each question names.
- Deep Link Drift
file -> Android package (APK). File type is the strongest signal -> Android RE. Name/theme "Deep Link Drift" + manifest meta-data huntress.mechanic = permissive scheme, host, and path validation -> the ...
- Half a Turn
"Thirteen steps forward" = ROT13 (rotate each letter 13 places). A 13-place rotation is its own inverse, so "funny how that works out" (apply it again to undo). Pure FAST PATH — no VM needed for the ...
- Host With the Most
file / 7z l -slt -> AES-256 encrypted zip. unzip (Info-ZIP) cannot do AES; used 7z. Theme "Host With the Most" + malware-sample packaging -> tried the malware-zip convention password infected first. Hit ...
- Looks Can Be Deceiving
Theme "Metadata Forensics" + title "Looks Can Be Deceiving" -> the note text is almost certainly a decoy; the real channel is file bytes/metadata. This is the Day 02 sibling of Day 01 "You Must Be This Hash ...
- Mangled
RULE 2 (file type): .zip -> carving/crypto. zipinfo -v showed AES encryption, so a password was needed before anything else. RULE 3 (theme): "Mangled" = C++ name mangling. The two members day01-c.elf (C) / ...
- BOM Voyage
Answer is a pure function of the file's first 3 bytes: EF BB BF -> true, else false. Single Solve and Time Trial: one zip, one file; false accepted both times (roster/evidence-note had no BOM). Mastery: two ...
- Expansion Pack
Each attempt ships an AES zip (password infected) with 3 PS1, 9 JS, 9 CMD/BAT; mastery ships two such sets (ids prefixed set1./set2.). PS1: base64 literal -> JSON. JS: run only inspect_() in node with eval ...
- Hex Appeal
Hint parsing: "Sixteen fingers" directly points to hexadecimal encoding Dynamic artifacts: Each mode has different assigned bytes; answers differ but format is consistent Multi-encoding: Mastery uses ...
- Intent Injection
The assigned APK contains assets/huntress_puzzle.json: {"schema":"huntress.mobile.puzzle.v1","day":3,"codec":"base64","data":"..."}. Base64-decoding data yields the backend answer code directly (the ...
- Table Manners
Forensic-inventory questionnaire, same family as Day 01 "ELF Aware" and Day 02 "Mangled"; d01/d02 were the already-known public samples (same sha256 as Mangled), d03 is new.
- Dim and Dimmer
Known family script_inventory (KB match). Reused challenges/_lib/solve_script_inventory.py for PS1/JS/BAT (159 fields resolved unchanged) and added only the VBS layer (72 fields).
- Go Fish
Same family as Day 03 "Table Manners" (dual-format RE inventory); new layer = the Go ELF. KB match: dual_format_re. Reused the banked solver, extended only d04.
- Pending Intent Relay
Mobile/Android, tiny synthetic fixture APK. Simplest-first: unzip, read the assets, strings on the dex, parse manifest strings. No apktool/jadx installed; not needed for an APK this small.
- Picture This
Metadata forensics on a JPEG. The relevant ground truth is the SOF (Start Of Frame) segment, which holds the real pixel dimensions regardless of filename or any EXIF/JFIF claims. Solver: scripts/sof_dims.py ...
- Two of a Kind
Warmup decode family (KB: warmup_decode). Single Solve artifact was one line of space-separated 8-bit binary decoding to Nice work decoding this warmup! To solve, submit this exact answer to the Game Master: .
- Backup Spill
Per the established pattern for these fixture APKs: check assets/.json first before any decompilation - the dex is small and decorative, the real answer lives in the bundled puzzle asset.
- By the Numbers
The prompt "characters have been reduced to numbers" directly names the encoding: each character of the message is represented as its decimal ASCII code, space-separated.
- Chunk Norris
Same family as Day 04's "Picture This" (JPEG SOF dimensions), but for PNG: the prompt explicitly distrusts rendering/declared metadata and asks to verify geometry from the file's own structure. For PNG the ...
- Hatch Careful
scripts/solve_hatch.py - one automated run per mode: start -> download -> extract -> analyze (6 languages) -> build 357/714 answers -> single batch submit. Handles mastery's multi-set layout (one zip per ...
- Rust Belt
GOTCHA (cost me 3 mastery attempts before the fix stuck across mutations): large ELFs contain MULTIPLE HCFG byte matches (decoys / false hits). Do not take the first data.find(b"HCFG") - scan every ...
- Eight Is Enough
Reconstructed from the saved solver (scripts/solve.py) - no separate recon notes or digest were kept locally, so this writeup documents what the script itself does, which is the only surviving record of the ...
- Pool Party
Not recorded. No solver script, writeup, or notes were saved locally before agent-3 moved on to its next challenge. This stub exists only to record the known facts above honestly - it is not a ...
- Portable Deception
Not recorded. No solver script, writeup, or notes were saved locally before agent-2 moved on to its next challenge. This stub exists only to record the known facts above honestly - it is not a ...
- Provider Paths
All three modes are CORRECT on the Game Master (state.json: done, modes single, time_trial, mastery). What the solve notes recorded (kb/lessons.jsonl, android_puzzle family): every mode uses the same ...
- Shell We Begin
Known family script_inventory (Day 01-05): per-field answers derived from fixture bytes. New layer this day: Bash.
- Node by Node
scripts/solve_node_by_node.py (reusable copy in _lib/) + new _lib/decode_node.py. Extends the Day06 unified solver (ps1/js/bat/vbs/py/sh) with an .mjs branch and advanced-bash feature/primitive handling. ...
- Percent Chance
Reconstructed from the saved solver (scripts/solve.py) - the title is the pun: "Percent" = URL/percent-encoding (%xx escapes), which is the actual transform being undone:
- Routine Check
Extracted the zip, hashed every file. Confirmed types via file/readelf. d01 ELF: readelf -h/-S/-s, objdump -d -M intel. recover_code (FUNC, 6 bytes @ 0x401126) = mov eax,0x1fa4 (8100) + ret -> disassembly ...
- Seven-Zip, No Relation
Reconstructed from the saved scripts (scripts/peek.py, scripts/run.py, scripts/submit.py) - the title is the pun: despite the "Seven-Zip" name this has "no relation" to the actual 7-Zip program/format; it's ...
- WebView Bridge
Reconstructed from state.json and the saved solver (scripts/solve_webview_bridge.py):
- Cleartext Session
Mobile-android lane. The claim-time per-lane autosolver (autosolve_mobile.compute) solved Single Solve directly from the APK bytes over the bearer GM, returning a backend answer code in the ANS- shape. ...
- Entity Relations
Fingerprinted instantly against the warmup_decode family in challenges/_lib/kb/families.json (not known-hard, Sonnet-medium base tier - borne out: this challenge solved in seconds). The title itself is the ...
- Lost in Translation
Day 08 extends the Day 02/03/06/07 script-inventory family (PS1 -> +JScript/batch -> +Bash -> +Node/Python-packers) with cross-language IOC normalization. 615 per-field answers (Single Solve / Time Trial) ...
- Swift Justice
Fingerprinted against the dual_format_re family in challenges/_lib/kb/families.json (known-hard; challenges/_lib/autosolve_binary.py is the banked day-generic adapter). The new, Day-08-specific layer: an ...
- Whats in the Box
File-type-driven metadata extraction, one field per answer_id. The only new work over prior days is the nested ZIP: parse its End Of Central Directory record for the total entry count and list its ...
- All the World's a Stager
Script-inventory family (the recurring PS/JS/batch/bash/Node multi-field inventory). The Day-09 new layer is stage relationships: each fixture declares a predecessor/successor family tree, and the task is ...
- Built Different
Metadata-bytes family (see docs memory ctf-evidence-note-decoy-family): every answer is a function of the assigned file's raw bytes, and extensions are untrusted - the drop mixes real executables for ...
- Escape Room
Warmup-decode family. The title is the hint: the day's new layer is escape sequences - the "characters would like to escape" means the blob is an escaped encoding (e.g. backslash/hex/unicode/HTML escapes) ...
- Nimble
Single Solve is untimed, so it was the lab: the two open fields were probed one id per submit (scripts/probe_d09.py + scripts/cands.json) with the candidate list ordered by what the bytes actually ...
- Trust Everyone
Android-puzzle family: the APK ships a permissive certificate-trust implementation (a custom TrustManager/network-security-config that accepts a challenge-planted CA). The answer is recovered from the APK's ...
- Bind Over Matter
Unzip evidence.zip with password infected (WinZip-AES; 7z x -pinfected / pyzipper). For each fixture, decode its Material_/Codec_ pair to the v=1| descriptor and read the keyed fields (role, attack ...
- Network Config Split
Each APK carries assets/huntress_puzzle.json (schema huntress.mobile.puzzle.v1). For this challenge the codec is split: data is empty and the answer is spread over parts, reassembled using the order index ...
- Thirty-Two Below
Pun: "Thirty-Two" = Base32 ("Smaller alphabet" than Base64). Artifact: blob.txt, a single Base32 layer (decode_chain.py: base32). Decoded text: "Nice work decoding this warmup! To solve, submit this exact ...
- Where Do We Begin
Correction to the first draft: this file recorded Time Trial / Mastery as "returned 409 timed_intent_not_available; rerun when they open". They were not gated. On a LEGACY single-pool timed route the GM ...
- Zig Zag
claim_next.py autosolved the single solve to 99/100 at claim. Identified the lone unresolved field (d10_analysis_has_hcfg_section) via analyse(); added the .hcfg-section handler => single_solve 100/100 ...
- By Any Other Name
A name isn't a behavior. The delivery carries 811 answer fields: three named PowerShell samples (resource-sample.ps1, collection-sample.ps1, impact-sample.ps1, one per MITRE tactic category - resource ...
- D-Day
Sequence, measured from the GM (not inferred):
- Local Secrets
Unzip the APK (6 members: manifest, classes.dex, certificate, assets). Read assets/huntress_puzzle.json. Today's variant: schema huntress.mobile.puzzle.v1, day 11, codec "xor", a base64 data string and an ...
- Safe Travels
Prompt: "Our usual base went on a URL-friendly vacation."
- Whats My Name Again
Filenames on disk are randomized; the organiser's own prompt is explicit that the recovery comes from the Windows PE VERSION_INFO resource instead. This attempt's delivery is one PE file, on disk as ...
- AutoIt Takes Two
autosolve_script.py line 44 already mapped the AutoIt layer before the release:
- Back at You
Layer: the blob is plain text written in reverse (character order reversed). Reversing it gives the sentence "Nice work decoding this warmup! To solve, submit this exact answer to the Game Master: ". The ...
- Managed Expectations
Question: Determine dotnet.mvid_guid for managed-001.dll (answer id d12-dotnet-001-dotnet-mvid-guid, GUID 8-4-4-4-12).
- Pascal's Wager
Day 12 is the twelfth rotation of the lane's single stable schema. The toolchain changes every day (C, C++, .NET, Go, Rust, Java, Kotlin, Swift, Nim, Zig, D, and today Delphi/Pascal - the title's wager) ...
- SQLite Assumptions
Unzipped the APK (artifact_0.bin): AndroidManifest.xml, classes.dex, META-INF signing files, assets/huntress_puzzle.json. assets/huntress_puzzle.json = ...
- Crystal Clear
claim_next.py waited for release, claimed the lane, and ran the banked lane adapter. Adapter path: ctf_gm.unzip (AES zip) -> classify each member by its bytes, not its extension -> binary_facts.facts_for ...
- Keystore Misuse
The artifact is an APK (6 entries: manifest, classes.dex, a DER certificate, text files, assets/huntress_puzzle.json). assets/huntress_puzzle.json = ...
- Opposite Day
artifact_0.bin is Atbash text: "Nice work decoding this warmup! To solve, submit this exact answer to the Game Master: handy-candle-quartz-31". AUTOSOLVE got both modes CORRECT. single_solve flag ...
- Product Placement
artifact_0.bin is a zip holding installer-001.msi (OLE2 compound file). MSI streams use an encoded stream-name alphabet; the Property table is column-major, with string references into _StringPool / ...
- Zip Code
claim_next.py claimed the challenge at release and ran the banked lane adapter (challenges/_lib/autosolve_script.py) before any human-style analysis. The adapter opens the AES zip with infected ...
- Predictable Crypto
The APK (artifact_0.bin, 6 members) carries assets/huntress_puzzle.json: {"schema":"huntress.mobile.puzzle.v1","day":14,"codec":"caesar","data":"JW\\6L>NN:=?","key":9} Today's codec is caesar with key 9. ...
- Selective Memory
GM prompt: "Objective-C never forgets: every class, every selector, every string is registered and waiting. Read the persistence descriptor and learn what this binary remembers."
- Seven Steps
Blob: Upjl dvyr kljvkpun aopz dhytbw! ... = Caesar +7 (decode -7) -> "Nice work decoding this warmup! To solve, submit this exact answer to the Game Master: dapper-sprout-boulder-71". Flag ...
- The Shortcut Home
artifact_0.bin is a zip holding one Windows shell link (shortcut-001.lnk, 547 bytes). Parsed the MS-SHLLINK structure by hand: header flags 0x97 (HasLinkTargetIDList, HasLinkInfo, HasName, HasWorkingDir ...
- Trust the Process
Day 14 adds MSBuild project carriers (.proj) to the cumulative script-triage inventory - the carrier that lane_progress.py had listed as a prepared next-layer candidate (msbuild project [PREPARED]). Four ...
- A Word to the Wise
artifact_0.bin is a zip holding document-15-001.docx (an OOXML package, itself a zip). Internal relationships (_rels/.rels) point at word/document.xml and docProps/core.xml (no TargetMode). ...
- Crumb Trail
Solve locally, submit once. If a timed batch comes back incorrect with no counts, do not re-analyse and do not start a fresh variant. Check the spent attempt's start response for deadline: null. If it is ...
- Native Library Secrets
Derive from each attempt's own APK bytes (puzzle is per-attempt, never hard-coded); one extractor serves all three modes.
- Room Forty-Seven
Flag: 6da8cecd2617866eb0e8190ae5fa2b0b Blob (107 bytes) is ROT47. Decoded: "Nice work decoding this warmup! To solve, submit this exact answer to the Game Master: sunny-castle-opal-96" Answer ...
- Stripped
All three modes landed on the first submit each (single 5.2 s, time trial 6.3 s, mastery 17.3 s), with 150 / 150 / 750 fields resolved locally and zero probe or unknown fields.
- Dot Com
Artifact is Morse for hex, hex decodes to "Nice work decoding this warmup! To solve, submit this exact answer to the Game Master: silly-banner-marble-88". Answer: silly-banner-marble-88. GM: CORRECT, flag ...
- JNI Boundary
artifact_0.bin is an APK (7 members: manifest, classes.dex, lib/x86_64/libarena.so, assets/huntress_puzzle.json, META-INF signing files). assets/huntress_puzzle.json = ...
- Return to Sender
Single Solve: CORRECT, flag 1783c21f7e23d03cf466847c903c28f5 Time Trial: CORRECT, flag 2d8fbf780074e7464a44f594ffbeb392 Mastery: CORRECT, flag 754302736d3c4de2ecae53c3a438bb1b
- Same Same, But Different
claim_next.py claimed the challenge and ran the banked lane adapter challenges/_lib/autosolve_script.py before any hand analysis. The adapter peeled the AES zip (pyzipper, password infected), walked all 123 ...
- Wasm Up
The banked lane adapter (challenges/_lib/autosolve_binary.py) answered 156 of 160 fields at claim time and correctly HELD the rest rather than guessing. The residue was exactly four ids, all of them the new ...
- Bacon Bits
Artifact: groups of 5 a/b -> 0..15 -> pairs of nibbles -> ASCII "Nice work decoding this warmup! To solve, submit this exact answer to the Game Master: tidy-rocket-harbor-20". Answer: tidy-rocket-harbor-20. ...
- Decoder Ring
Mastery returns zero per-field feedback on an incorrect whole submit. But a partial submit that is 100% correct for every included field returns accepted with accepted_ids and remaining. So the oracle is ...
- Fetch Quest
No new reversing was needed. The banked lane spine (challenges/_lib/autosolve_script.py, driven by challenges/_lib/autosolve.py) handled the whole thing:
- Hear Me Out
The actual codec comes from the bytes (fmt chunk format tag 1, 16-bit, mono, 16 kHz), not the name: pcm_s16le (ffprobe codec_name spelling). New extractor challenges/_lib/media_codec.py (ffprobe first ...
- Name Collapse
Single, Time Trial and Mastery all CORRECT via autosolve (Single 1.1s, TT 3.5s, Mastery 2.6s). Single answer (decoded from the shipped APK): ANS-E26A5C21D3110FA1
- Encrypted Strings
artifact_0.bin is an APK (zip, 6 members): AndroidManifest.xml, classes.dex, assets/huntress_puzzle.json, META-INF/. assets/huntress_puzzle.json = ...
- Flow State
Implemented the GM's rule literally, so the fields answer SURE with no candidate list:
- Frame of Reference
Single Solve, Time Trial and Mastery all CORRECT (accepted by the GM).
- Off the Rails
Blob is the carrier sentence ("Nice work decoding this warmup! To solve, submit this exact answer to the Game Master: ") transposed with a 3-rail rail fence. Decoding with k=3 recovers it. AUTOSOLVE ...
- Orphan Process
Solved by the team solver. Single 1194/1194 fields accepted; Time Trial and Mastery solved from their own mutated bytes by the same extractor (no per-mode rework, no timed start by hand - AUTOSOLVE only).
- Back Up a Minute
gm.start(challenge) returns one live target site per question (proof-1 for Single Solve/Time Trial; proof-1+proof-2 for Mastery, one target each). Target site: a small Node.js "Status Sprout" status page. ...
- Jack in the Box
Solved by the team solver. Mastery ran as a RANKED attempt inside a 324 s server window (deadline 2026-10-19T13:27:34.585Z) and answered 1140 fields in 15.3 s with 0 probes and 0 unknowns.
- Little Endian on the Prairie
Artifact is a zip with one member, collector-001.elf. The answer comes from the ELF header bytes, not from the filename or any note: ELF64 (EI_CLASS=2), little endian (EI_DATA=1), so e_machine is the u16 at ...
- Little Things
Single Solve: CORRECT (autosolve_warmups, 1.9s) Time Trial: CORRECT (6.4s) Mastery: CORRECT (2.8s, ranked)
- Permission to Slip
Up to Day 19, the web lane had NEVER staged live - all web coverage was synthetic offline against the four web_contracts.py executors (hash, jwt, config, http). None of them covers: an id-bearing JSON API ...
- Query Me This
Day 3 - Web Exploitation - boolean-blind SQL injection in a search box - all three modes CORRECT on 2026-10-19 via the homestation Commander (same backlog pickup as Permission to Slip; agent-6 did not claim ...
- Reflection Maze
artifact_0.bin is an APK (6 members: AndroidManifest.xml, classes.dex, assets/huntress_puzzle.json, META-INF signature files). assets/huntress_puzzle.json = ...
- Sharp Practice
Each timed mode was opened only by AUTOSOLVE after its readiness gate passed (prior mode already CORRECT live on the same challenge, so same compute was proven on real bytes). No mode was started, retried ...
- Administratively Speaking
Each mode gives one or more live target URLs under .2026.huntress.ctf.games. No downloadable artifact. The question in every mode is "Recover the proof from Target N", answer kind exact_text, normalization ...
- Big Entrance
single_solve: CORRECT (autosolve, 1.2s) time_trial: CORRECT (autosolve, 3.0s, ranked) mastery: CORRECT (autosolve, 2.9s, ranked)
- Loaded Later
artifact_0.bin is an APK (7 members): AndroidManifest.xml, classes.dex, assets/huntress_puzzle.json, assets/payload.dex, META-INF/FIXTURE-.{SF,RSA}, MANIFEST.MF. assets/huntress_puzzle.json = ...
- Resourceful
Every mode answered 100% SURE: 240 sure, 0 probe, 0 unknown, then 1200 sure, 0 probe, 0 unknown.
- Stay in Your Directory
Both the Single Solve and Time Trial targets are a PHP app ("Runbook Nook" / "Stay in Your Directory") where a document viewer includes files relative to its own pages directory:
- The Butler Did It
Standard script-triage universe pattern - every fixture is a self-describing inert specimen whose first line is a HUNTRESS_ANALYSIS_V1:HEX- descriptor and whose body re-materialises the same hex-encoded ...
- Universal Appeal
Flag: 29f836971748707c3e26a94cab556c38 artifact_0.bin is a zip with one member, macos-001.bin (56 bytes): a thin little-endian 64-bit Mach-O (cffaedfe), cputype bytes 0c000001 = 0x0100000c (arm64), one ...
- Actions Speak Louder
RECOGNIZE -> the claim digest fingerprinted the artifact as the lane's known Material_/Codec_ self-decoding fixture family (TWIST constant since Day 6).
- Debug Me If You Can
The banked lane autosolver (challenges/_lib/autosolve_mobile.py) ran at claim time, before this agent saw the claim. It reads assets/huntress_puzzle.json from the APK, decodes the data field with the codec ...
- Package Deal
The artifact is a zip holding one member, mobile-001.apk (8535 bytes). The apk is itself a zip: AndroidManifest.xml, classes.dex, META-INF/FIXTURE-.SF/.RSA/MANIFEST.MF. AndroidManifest.xml is binary XML ...
- Reflect Upon This
GET / shows a "Search Mirror" Fastify app. GET /api/clues returns the feature route (/execute-...), the callback route (/callback-...) and a per-target nonce. The feature route reflects the q parameter as ...
- Session Impression
scratchpad/si_solve.py runs the three-step attack for every target returned by the Session Impression start, maps answer_ids proof-N to targets by order, and submits in one batch. It worked for Single and ...
- Shared Custody
Two new evidence-driven member selectors, applied in answer() only when the id carried no member tag and the dNN really has two or more binary members (artifact_sha256_N is never re-pointed - it is numbered ...
- Stored for Later
solved_by on all three is the shared team identity the team solver (expected).
- Under Pressure
The warmups adapter (challenges/_lib/autosolve_warmups.py) walks each set's lines against a fixed positional decoder ladder in LADDER_NAMES. Line 1 is the plaintext; lines 2..n are its encoded forms, in a ...
- Bridge Troll
Single landed on the second submit (231 fields accepted first pass, then the two adapter fixes below were applied and the remaining 33 fields went in; 264/264 at 7.6 s on the second run). Time Trial and ...
- Deflated Ego
Recognize: warmup family, one encoded blob peeled until the carrier sentence appears; answer is the word-word-word-NN slug in the final sentence. The title hints at a deflate-style compression layer. ...
- Forged With Care
Send the assigned target a cross-site GET with the active session cookie. Set the destination to the external review mailbox and omit csrf_token:
- Main Character Energy
Recognize: metadata-forensics family, "Determine . for " question shape. The field key names the extractor. Bind/Extract: the autosolver read the answer from the claimed artifact's raw bytes using the ...
- Next Next Finish
The project's banked lane adapter challenges/_lib/autosolve_script.py already covers this family (script_inventory, generic across days). I did not write a per-day solver.
- Root of Trust
Theme: certificate pinning / permissive CA trust (cosmetic). The real layer is the puzzle codec in assets/huntress_puzzle.json inside each APK; decode data with the codec it names -> ANS-. The banked ...
- Uploading Expectations
The GM supplies live targets in each mode’s start response. Each target was handled independently; the Mastery targets used three different web paths despite sharing the same challenge prompt.
- Ask Again Later
solved_by on all three is the shared team identity the team solver.
- Eighty-Five Alive
Recognize: the title carries the hint "85", which points to a base85 layer. This follows the lane pattern of one encoding per day, with the answer being the three-word-two-digit slug in the final sentence. ...
- Hook Line and Signal
The puzzle JSON carried no data at all (schema huntress.mobile.puzzle.v2, day, codec: analysis, build). The answer is not stored anywhere; it is computed by the APK's own code: Release.code() = ...
- Page Me Maybe
One answer id: d23-sqlite-001-sqlite-page-size-bytes. The staged database-001.sqlite reads 4096 at offset 16-17 (u16 BE page size). AUTOSOLVE submitted 4096 in 1.9 s.
- Posted Too Much
The target is a NestJS/Express app. /api/clues is the per-attempt route map:
- Press Start to Decode
Each fixture carries its own self-describing descriptor; the only new thing on Day 23 is the richer set of codecs and the file-naming scheme (. with no polyglot-- prefix).
- Template Temper
Answer format for every target: exact_text, normalization trim — submit the verbatim proof_ string recovered from each target. Flags/proofs above are the GM-accepted values from ...
- Trust Issues
Pipeline flow (all three modes solved by the team solver): Single Solve: CORRECT in 11.7 s (untimed, static artifact). Time Trial: ranked, EARLY start (prior Single CORRECT live), CORRECT in 8.3 s. Mastery ...
- Emulator Alibi
All three modes CORRECT: single_solve, time_trial (ranked), mastery (8 APKs, ranked attempt window).
- Event Horizon
Solved by the banked metadata-forensics lane autosolver (autosolve_metadata, capability layer), run as autosolve.py 24 metadata-forensics-day-24-event-horizon metadata-forensics. It derives each answer from ...
- Fetch Quest
Mastery is the real test: each of the 4 targets is a different web vuln bundled under the SSRF "Fetch Quest" theme. The Single and Time Trial variants are the same 4 archetypes, drawn in isolation.
- Spot the Difference
Patch-diffing challenge in the binary-reverse-engineering lane. The lane's banked autosolver (challenges/_lib/autosolve_binary.py, routed via the generic autosolve.py with the team solver as the solver) ...
- String Theory
Slug: jaunty-compass-mosaic-76. Derived locally from the saved artifact, then submitted through autosolve. Flag: e19f67f33461d0495f9668cc6199061d
- The API Has Opinions
Opened the Single target, read /, saw the hint "Open workspace notes" → /api/clues. /api/clues exposed feature_route, protected_reference; /api/inventory listed the field/action vocabulary. Enumerated (id ...
- Web of Lies
Prompt framed the artifact as a JSP "webshell" source that reads requests and stages server-side work; the task is to recover the request/stage order and separate inert fixture behavior from lines that ...
- Ashes to Assets
AES zip evidence.zip (password infected), 167 inert fixtures in ~26 languages (ps1, js, vbs, bat, py, sh, php polyglots, Elixir/Phoenix host, JSP, Lua, Ruby, ...). Every fixture holds a Material_ / Codec_ ...
- Hive Mind
Prompt: registry hives from multiple systems were renamed; identify each true root, regroup the systems, isolate the planted hive.
- Integrity Signals
RECOGNIZE: the artifact is an APK with assets/huntress_puzzle.json (the lane's known puzzle mechanism). BIND/EXTRACT: autosolve_mobile read the asset statically from the APK. Single used the hex codec; Time ...
- Out of Bounds
The d25 members are a two-member JAR pair distinguished only by a basename role suffix:
- Reset Expectations
Each target advertises its own routes: GET /api/clues -> {feature_route (submit), secondary_route (result), recovery_guide_route} GET /api/recovery -> {request_route, redeem_route, account_parameter ...
- Resolver Revolt
caseIndex labels one id as "reference only", but the case(id) resolver has NO visibility check: it returns the note for that id anyway. Query:
- Shell Company
/api/clues advertises feature_route (e.g. /process-38db5f), input_parameter (path), example_path (/app/static). The parameter is passed straight into a shell command (ls -ld $path), so ;/&&/| inject. ...
- You Have a Point
Single Solve: answer brave-beacon-wizard-34 -> CORRECT, flag 46f599fcae691c17b1a86497d9744e0c Time Trial (ranked): answer jaunty-beacon-clover-24 -> CORRECT, flag 63e2f47cb55303bbe2d3a284ba23ee91 Mastery ...
- Before You Go
Description: Prefetch evidence records what Windows prepared to execute even after filenames changed. Recover executable identities and connect execution traces to the correct hosts.
- Crash Course
"Crash Course" is the Day 26 instalment of the recurring cumulative binary-config extraction mastery. One encrypted evidence.zip (AES-256, password infected) carries one deliverable per day (dNN--...). Each ...
- Food for Thought
All three modes completed: Single Solve, regular Time Trial, and regular Mastery. Each Game Master status response returned CORRECT.
- Mixed Signals
A scaled-up version of the Day-22 "installer-script-chain" script-triage family. Each mode is a batch of answer ids, each id derived from a script fixture inside an evidence archive. The theme ("Erlang ...
- Serialized Offender
Deterministic ladder, not an LLM guess. The whole family pool is now covered by challenges/_lib/web_proof_recon.recover(url) after the 2026-10-26 promotion (see §6). recover() walks the ladder and returns ...
- Split Decision
The artifact is an Android App Bundle delivered as a .apks container (an outer ZIP, magic PK\x03\x04) that holds two split APKs:
- Trust the Webhook
This section is the solve itself, recorded from the run on our Homebase VM. The sections after it are the working notes from the earlier live attempt, kept as written; the blocker they describe was resolved ...
- Compiler Says What
Prompt: every compiler signs its work - compare the runtime and linker fingerprints across builds and name the toolchain, attributing this binary to its maker.
- Dependency Surface
Prompt: prove which generated dependency flaw is reachable from an exported mobile surface.
- Double Spent
GET /api/clues to resolve the per-target feature/secondary routes. Fire ~40 parallel POST /execute- requests (no body, no auth, no cookies). GET /audit- -> the response body carries evidence: proof_. Submit ...
- Forwarded Misgivings
solver_copy.py additions: _tech_forwarded_misgivings: /api/clues → feature_route → scoped_get(..., headers={"X-Forwarded-For":"127.0.0.1"}) → extract evidence → return proof. Walks a short header ladder ...
- FROM Whence It Came
agent_claim.py agent-1 570 acquired the CID on first contact with the GM. agent_solve_cid.py --agent agent-1 --day 27 --cid --lane script-analysis ran the whole three-mode pipeline on the already-prepared ...
- Monkey Business
Ook tokens (Ook. / Ook? / Ook!) are read in PAIRS and map 1:1 onto brainfuck: .?=> ?.=< ..=+ !!=- !.=. .!=, !?=[ ?!=]. The Single blob (33929 bytes) held only Ook. and Ook! (3393 pairs: 1667 +, 1618 -, 108 ...
- Packet Loss Prevention
Lane solver autosolve_metadata (per-CID working copy scripts/solve.py, unmodified) dispatches on each question's text, binds the artifact member bytes and runs a byte-level metadata extractor (format.field ...
- Socket to Me
Target: https:/// (Koa app "Live Case Wire", role=analyst).
- Back in Sixteen
The row numbers identify different encodings. Rows 1–11 decode into the opening “Nice work decoding this warmup! To solve,” which confirms that the file is a message assembled from encoded fragments. The ...
- Fail Open Friday
GET /api/clues on the assigned target. The response identified the feature route and documented the valid public decision shape: remote:: with a documented standard decision. Sending a MALFORMED value with ...
- Family Resemblance
Started each mode through the Game Master proxy and saved the complete start response. Time Trial and Mastery used run_for_time: true. Downloaded each assigned evidence.zip through ?op=artifact. For ...
- If a Breach Falls in a Log
GET /api/clues to obtain the feature route + secondary audit route. POST {} to the feature route → response reports a completed protected_export and an export_id. GET the secondary (audit) route → find the ...
- Out of Order
Ground truth available without a GM: GM record .../_gm_record/...homestation.jsonl (13 entries): single start (1703 questions + answers), single submit → CORRECT flag 2bb8d7b14e84352dda4167b5b31112f5, and a ...
- Protocol Reconstruction
Single Solve, Time Trial, and Mastery were all accepted by the Game Master. The Time Trial and Mastery starts used the default practice pool (run_for_time:false), so the server returned no deadline and the ...
- Where It All Began
The supported proxy was used for every Game Master request. For example:
- Fraternal Twins
Only this challenge's own solver copy was patched; shared lane kits were left untouched. challenges/Day 29/⚙️ Fraternal Twins/scripts/binary_wip.py
- Internal Affairs
Target: SS Target 1 Attempt ID: [attempt-id] Answer ID: proof-1 Timer: no deadline
- Make It Make Sense
claim_next.py agent-1 -> single_0.bin pre-fetched; script_wip.py cloned as the per-CID solver copy. autosolve_script.run() iterated the ZIP members, peeled the carrier, ran solve_fields_ext on the decoded ...
- Rotten Base
Prompt: "Sixty-four invited thirteen over. Things got scrambled." -> base64 (sixty-four) with ROT13 (thirteen) applied over it.
- Trust Issues
Artifact: single_0.bin, 672 bytes, a ZIP (metadata-day-29-evidence.zip) holding one member, certificate-001.pem. Question: d29-x509-001-x509-sha256-fingerprint - "Determine x509.sha256_fingerprint for ...
- Trust the Build
Started through the Game Master with challenge ID web-exploitation-day-26-trust-the-build. The SS attempt ID was [id]. This is a web challenge, so the start response had no downloadable artifact; it ...
- White Box Key Recovery
All three modes CORRECT (GM-accepted flags):
- Campaign Trail
The lane spine already answered 413/420 of the Single's ids by catalog (d01..d29 shapes + d30 structural / artifact_sha256_N / Neko-based elf-less inventory). The 7 residues were the new d30 class:
- Layered Protections
APK with assets/huntress_puzzle.json: {"schema":"huntress.mobile.puzzle.v1","day":30,"codec":"layered","data":"TU49OyInJlUlJSQkKilcKDFOUF4=","key":12} Derivation (static): base64-decode data (20 bytes) -> ...
- Pass the Identity
The live target advertises itself at /api/clues:
- Percent Percent
Added LINE_DECODERS["percent2"] (unquote twice) and LADDER_NAMES_D30 = D29 + ["percent2"]. _guided_positional selects the D30 ladder when a blob has exactly 30 lines. Offline check against the 148 ...
- Queue the Confusion
GET /api/clues -> {feature_route, secondary_route, owner_directory_route, current_role: analyst}. GET /api/owners -> {caller: analyst, restricted_owner: executive}. POST with {"owner": "executive"} is ...
- Seeing Double
Hash every member from its real bytes (never the manifest claim), group equal digests, keep groups of 2+ members. Result: 5 groups of 6.
- SHA1-Hulud
Debugging was off the table from the second check (ThreadHideFromDebugger + VEH + DRx reads), so the whole arc was snapshot the real process after the anti-tamper loops, then continue in a Unicorn emulator ...
- Spill the Beans
/tmp/spill-mastery-status.json — full active-attempt status and question list. /tmp/spill-the-beans-mastery.zip — downloaded assigned Mastery archive. /tmp/spill-beans-sets/evidence-01 through evidence-14 — ...
- Final Boss
"The gauntlet: everything you've learned ... in one campaign." The artifact is an AES-256 evidence.zip (password infected, 46 members). It has one dNN-- member group for each of Days 01-31 (ELF, PE, .NET ...
- Framework Roulette
Result: all three modes CORRECT by 13:51 ET on release day (Single Solve, Time Trial, Mastery with six accepted ids). Time Trial and Mastery were regular untimed attempts, not ranked. All three submits ...
- Full Stack Incident
The target exposed its own clues and request log. Those endpoints were the route map; the route names were randomized for this instance, so they should be read from /api/clues instead of guessed or hardcoded.
- Mobile Arena Gauntlet
Every APK carries assets/huntress_puzzle.json with codec: "analysis" (schema huntress.mobile.puzzle.v2, a day and a build). The answer is not in the JSON: it is the String returned by Release.code(Context) ...
- No Artifact Left Behind
Only one field was new: graph.execution_chain_artifact_ids for case-001.zip. Every other id in the Time Trial (36) and Mastery (540) used extractors that already existed. The answer is a compact JSON array ...
- One More Thing
Single / Time Trial: one blob, base64 -> gzip -> base32 -> slug; solved by the stock decoder chain with no change. Mastery: 153 numbered-line blobs of 31 rows each (ladder D30 plus a new row 31). Rows 1-30 ...
- The Full Transform
Static only. The stock autosolve_script decodes every descriptor and answers role/attack/style/campaign/steps/url/ indicator/secondary/tertiary and the compound fields (campaign-token, file-path, path-leaf ...