BB
← All projectsBitBox

BitBox02 firmware

Firmware and bootloader for BitBox02 signing devices.

BitcoinHardware walletsNormal
Repository coverage

647 commits in the local evidence base

Every captured commit receives deterministic security triage and a separate communication-quality score. Security candidates and broader second-pass signals receive full-patch Ollama analysis.

193security candidates127second-pass queue103AI analyses
28commits · 30 days
47commits · 60 days
312commits · 180 days
647commits · 365 days
Backfill bands
Aug 5 → Feb 6335 seen28 candidatesComplete
Feb 6 → Jun 6265 seen19 candidatesComplete
Jun 6 → Jul 619 seen5 candidatesComplete
Jul 6 → Aug 526 seen3 candidatesComplete
Commit communication

Does the history explain itself?

Message quality measures whether a commit identifies its scope, purpose, rationale, testing, and supporting references. It does not change the security-severity score.

59/100 average clarity
65Strong · 80–100
281Adequate · 60–79
230Thin · 40–59
71Opaque · 0–39
23security candidates with opaque commit messaging
Read the scoring rubric →
Developer activity

Who is changing the project?

Public Git author strings; identities are not independently verified.

DeveloperCommitsCandidatesAnalyzedHigh riskMessage avg.
Marko Bencun39011765258
Niklas Dusenlund1112912059
benma's agent892816062
cedwies1253063
Tomas Vrba944074
Cedric Wiese1231049
Jad810062
thisconnect210072
benma210074
Niklas111035
Yasser Aziza111070
Patrick Steiger110045
Analysis record

Published AI watches

Last scanned 44 minutes ago

Low 39 AI analysisMessage 73 · Adequate
BB BitBoxBitBox02 firmware BitcoinHardware wallets

rust: initialize C output buffers

This commit fixes a class of low-level memory-safety bugs where Rust code was given buffers containing uninitialized bytes. Rust's rules require every byte of a slice to be initialized, even if the function will overwrite them. Passing uni…

Undefined behavior at C/Rust FFI due to uninitialized buffers being treated as Rust slicesPotential optimizer-dependent behavior from violating Rust slice initialization rulesHardening of cryptographic output paths (SHA-256, HMAC-SHA256, HMAC-SHA512)
511018eaby benma's agent+52−2713 files
Vendor flagged security relevance
Moderate 59 AI analysisMessage 69 · Adequate
BB BitBoxBitBox02 firmware BitcoinHardware wallets

Limit SD erase file size

This commit fixes a bug in how the BitBox02 hardware wallet wipes files from its SD card. Before erasing a file, the device now checks the file's reported size against a safe maximum. Without this check, a tampered SD card could claim a fi…

CVE-2026-6682 referenced in commit messageMalformed FAT directory entry could cause excessive overwrite loopDenial-of-service via SD card tampering
2453f528by Marko Bencun+4−01 file
Vendor flagged security relevance
High 70 AI analysisMessage 66 · Adequate
BB BitBoxBitBox02 firmware BitcoinHardware wallets

Validate mounted FAT geometry

This update adds a safety check when the BitBox02 hardware wallet mounts a microSD card. A malicious or deliberately malformed FAT filesystem could trick the device's file-system library into placing user data inside attacker-controlled bo…

Fixes integer-wrap / geometry confusion in FAT mount logicAdds explicit post-mount validation of filesystem metadataPrevents data area from landing inside attacker-controlled FAT sectors
01c017d6by Marko Bencun+21−01 file
Vendor flagged security relevance
Informational 0 AI analysisMessage 45 · Thin
BB BitBoxBitBox02 firmware BitcoinHardware wallets

blupgrade: update stage1 binaries to v1.2.2

This commit simply swaps in newer pre-built bootloader stage1 binary files (version 1.2.2 replacing 1.2.1) for four BitBox02 hardware variants and updates the corresponding checksum list. The actual code inside the new binary files is not …

5940a800by Marko Bencun+8−86 files
No security note in commit
Moderate 59 AI analysisMessage 50 · Thin
BB BitBoxBitBox02 firmware BitcoinHardware wallets

bootloader/stage1: fix erase handling for partially erased blocks

This update fixes the BitBox02 bootloader's firmware-erase routine. Previously, when erasing leftover padding after a firmware update, the bootloader started erasing at the exact page where the firmware ended. Because flash memory can only…

Bootloader firmware erase routine could erase a flash block containing both firmware and paddingFix aligns erase start to erase-block boundary and re-checks erased state before erasingChangelog describes the change as a fix for 'partially erased flash blocks'
b31206a8by Marko Bencun+23−83 files
No security note in commit
Informational 15 AI analysisMessage 45 · Thin
BB BitBoxBitBox02 firmware BitcoinHardware wallets

blupgrade: add stage0/stage1 production binaries

This commit adds production bootloader upgrade files for the BitBox02 hardware wallet and updates build scripts to use them. It is a routine asset-management change: replacing placeholder development hashes with real signed production bina…

8db4b0dcby Marko Bencun+26−1720 files
No security note in commit
Informational 20 AI analysisMessage 83 · Strong
BB BitBoxBitBox02 firmware BitcoinHardware wallets

blupgrade: keep dev stage1 unsigned

This commit fixes a build script used only for development/testing versions of the BitBox02 bootloader upgrade. It makes the development-stage1 bootloader images unsigned again, while keeping production images fully signature-verified. The…

Signature verification relaxed only for development buildsProduction payload validation still requires signaturesDevelopment stage0 already skipped stage1 signature verification per commit message
476b90e3by Marko Bencun+9−69 files
No security note in commit
High 76 AI analysisMessage 23 · Opaque
BB BitBoxBitBox02 firmware BitcoinHardware wallets

security improvements

This BitBox02 firmware update is a broad security patch that fixes several independent bugs: it prevents a maliciously oversized USB report from overflowing memory, stops a corrupted Bluetooth pairing database from being read or written wi…

Bounds check added to USB HID Set Report input lengthBLE bond DB length validation hardened against negative and oversized valuesBootloader firmware image size limit relaxed to intended maximum
cbb40634by Marko Bencun+1117−25021 files
Vendor flagged security relevance
Low 46 AI analysisMessage 60 · Adequate
BB BitBoxBitBox02 firmware BitcoinHardware wallets

bootloader: allow full sized images

This commit fixes a bootloader bug where the device rejected firmware updates that used the maximum allowed size. The off-by-one check meant legitimate full-sized firmware images could not be installed, potentially blocking updates. The fi…

Off-by-one input validation in firmware-update pathBootloader change affecting firmware chunk count acceptanceCHANGELOG labels the change as a bugfix for full-sized firmware upgrades
f60b93ccby Marko Bencun+5−33 files
No security note in commit
Moderate 59 AI analysisMessage 28 · Opaque
BB BitBoxBitBox02 firmware BitcoinHardware wallets

Add bootloader update

This is a large firmware commit that adds a new two-stage bootloader update mechanism for the BitBox02 hardware wallet. It replaces the old single bootloader with a small, fixed 'stage0' plus a separately signed 'stage1', and ships a speci…

Bootloader architecture changed from monolithic to two-stage (stage0 + signed stage1).Firmware signature hash now includes a 16-bit product_id, binding firmware to product variant.Root public keys were rotated/replaced with a single set across all products.
3f1f3172by Marko Bencun+5003−52379 files
Vendor flagged security relevance
Informational 17 AI analysisMessage 78 · Adequate
BB BitBoxBitBox02 firmware BitcoinHardware wallets

Bump Rust toolchain to 1.96

This commit is a routine maintenance update for the BitBox02 hardware wallet firmware. It upgrades the pinned Rust compiler toolchain from an older version to Rust 1.96.0, updates the project's Docker build container and related developer …

Routine toolchain and dependency upgradeVendored libc updated across many platform bindingsNo first-party security-relevant code changes
55e3dd30by Niklas Dusenlund+14839−13798360 files
No security note in commit
Informational 18 AI analysisMessage 68 · Adequate
BB BitBoxBitBox02 firmware BitcoinHardware wallets

bootloader: reduce snprintf

This commit removes the use of the standard snprintf formatting function from the BitBox02 bootloader's developer menu and replaces it with a small, fixed-format helper. The main goal is to reduce the bootloader's binary size and make the …

Removes snprintf/printf usage from bootloader code, reducing binary bloat and eliminating a complex formatting library from a privileged execution contextExtends automated CI check to detect stdio/Rust formatting symbols in regular development and production bootloader outputs, not just production buildsAdds unit test coverage for the new fixed-format helper
2fb9d57fby Marko Bencun+59−106 files
No security note in commit
Low 35 AI analysisMessage 63 · Adequate
BB BitBoxBitBox02 firmware BitcoinHardware wallets

bootloader: avoid snprintf formatting

This commit removes the use of standard string-formatting functions (like snprintf and Rust's format!) from the bootloader and related low-level code, replacing them with smaller, fixed-size string builders. The stated goal is to reduce bi…

Removal of stdio formatting (snprintf family) from bootloader codeRemoval of Rust format!/write! formatting from bootloader-linked codeAddition of CI gate rejecting stdio and Rust formatting symbols in production bootloader ELFs
33434fa9by Marko Bencun+298−6314 files
No security note in commit
Informational 15 AI analysisMessage 35 · Opaque
BB BitBoxBitBox02 firmware BitcoinHardware wallets

bootloader: extract formatting helpers

This commit is a simple code cleanup: it moves a handful of string-formatting helpers (for pairing codes, progress percentages, hash display, timers, and unknown-command messages) out of the main bootloader file into a new dedicated file, …

1a1de3b5by Marko Bencun+147−126 files
No security note in commit
Informational 17 AI analysisMessage 58 · Thin
BB BitBoxBitBox02 firmware BitcoinHardware wallets

bootloader: render messages directly

This commit removes a 100-character stack buffer and passes bootloader messages straight to the screen-drawing function. The old code used snprintf to copy the message into a fixed-size buffer, which could silently truncate very long strin…

Removal of fixed-size stack buffer in bootloader display pathElimination of snprintf with potentially attacker-influenced format string argumentBootloader code touched, which is a security-sensitive component
c0bc7b92by Marko Bencun+1−31 file
No security note in commit
Informational 19 AI analysisMessage 58 · Thin
BB BitBoxBitBox02 firmware BitcoinHardware wallets

Use integer progress fractions

This commit replaces floating-point progress calculations with integer math to save firmware space. It removes the use of f32 arithmetic for progress bars in Bitcoin transaction signing and Bluetooth firmware upgrades. There is no direct s…

Removal of floating-point arithmetic reduces attack surface related to soft-float emulation bugsNew integer fraction API introduces denominator-non-zero and numerator-bound assumptionsCall sites use checked arithmetic to prevent overflow in progress computation
17c4947fby Marko Bencun+52−3712 files
No security note in commit
Informational 12 AI analysisMessage 80 · Strong
BB BitBoxBitBox02 firmware BitcoinHardware wallets

Enable scope C LTO for firmware

This commit changes the build system to enable Link-Time Optimization (LTO), a compiler technique that shrinks the final firmware size. It also adds safeguards to keep the stack protector security feature working under LTO and deliberately…

Build-system change enabling LTO with explicit stack-protector symbol retentionDeliberate exclusion of bootloader, factory-setup, ASF4, samd51a-ds, and embedded-swd from LTO due to audit complexityToolchain switch to LTO-aware archive utilities to avoid invalid tiny images from missing live objects
0b102fb7by Marko Bencun+29−36 files
No security note in commit
Moderate 60 AI analysisMessage 95 · Strong
BB BitBoxBitBox02 firmware BitcoinHardware wallets

da14531: restore BLE UART backpressure

This update fixes a crash in the BitBox02 hardware wallet when a user lists many backups over a Bluetooth connection. Previously, the device could run out of temporary buffer space while sending a large response, causing it to panic and st…

Denial-of-service via buffer overflow/panic on Bluetooth pathMissing backpressure leading to unbounded queue growth and crashAtomic all-or-nothing enqueue to preserve serial frame boundaries
3201c5d2by Marko Bencun+87−37 files
Vendor flagged security relevance
Informational 20 AI analysisMessage 35 · Opaque
BB BitBoxBitBox02 firmware BitcoinHardware wallets

securechip: port attestation sign

This commit is a code refactoring that moves the device attestation signing function from C code into Rust code. It converts the function to be asynchronous (async/await) and rewrites the parsing of the secure chip's DER-encoded signature …

Removal of C synchronous secure chip signing wrapperAddition of async Rust secure chip ECDSA signing operationRelocation of DER signature parsing from C-API Rust crate into securechip crate
2bccdebfby Marko Bencun+180−14525 files
No security note in commit
Informational 15 AI analysisMessage 58 · Thin
BB BitBoxBitBox02 firmware BitcoinHardware wallets

Format generated protobufs explicitly

This commit is purely a code-style and build-maintenance change. It makes the Rust code generated from protocol-buffer definitions be formatted by rustfmt automatically and included in style checks. No security-sensitive behavior is change…

f22f12cdby Niklas Dusenlund+36−1603 files
No security note in commit
Repository ledger

Explore captured commits

Expand any commit for its author, full message, clarity score, changed files, triage signals, analysis, and source link.

Security candidatehal: add random, factory_randomness, drop random_32_bytesby Marko Bencun · 47454d61 · Apr 8, 2026 · 27 filesMessage 55 · ThinLow 45Details
Commit message · Marko Bencun

hal: add random, factory_randomness, drop random_32_bytes

55/100 · ThinMessage clarity
✓ Specific, descriptive subject✓ Names a concrete action or component✓ Names security-relevant behavior explicitly! No meaningful explanatory body
Why it was queued
seed or entropy pathsigning or wallet path
AI analysis · Low 45/100

This commit changes how the BitBox02 hardware wallet generates random numbers used to create secret keys and encrypt stored data. Previously, the firmware used a single 32-byte random value from the microcontroller. After this change, it mixes three sources: the microcontroller's random generator, the secure chip's random generator, and a fixed 32-byte 'factory randomness' value stored in a special flash memory location. The commit also adds a factory script to write that fixed value into devices during production. The change is a defensive hardening measure, not a fix for an active bug, but it introduces a new dependency: if the factory randomness is not actually random or is reused across devices, it could weaken security instead of strengthening it.

Security candidateeth: harmonize large data warnings for transactions and typed messagesby Tomas Vrba · 56a6ec34 · Apr 7, 2026 · 5 filesMessage 73 · AdequateLow 35Details
Commit message · Tomas Vrba

eth: harmonize large data warnings for transactions and typed messages

Adjust wording in transaction signing so warnings and truncating logic
match the newly implemented approach in sign_typed_msg.rs

Co-Authored-By: Marko Bencun <marko@shiftcrypto.ch>

73/100 · AdequateMessage clarity
✓ Specific, descriptive subject✓ Names a concrete action or component✓ Provides detailed explanatory context
Why it was queued
signing boundarysigning or wallet path
AI analysis · Low 35/100

This commit changes how the BitBox02 hardware wallet warns users about very large Ethereum transaction data and typed messages. Previously, large transaction data was shown only as a byte count with the message 'too large to display.' Now the device shows a preview of the actual hex data (up to the screen limit, with a '...' truncation indicator) and adds a warning screen saying the value is too large to display in full. The change also reuses the same truncation logic already used for EIP-712 typed messages, so both flows behave consistently. The commit does not fix a vulnerability, but it improves the user's ability to inspect what they are signing.

Security candidaterust: async unit tests instead of block_onby Marko Bencun · 2560d5d1 · Apr 6, 2026 · 41 filesMessage 55 · ThinInformational 15Details
Commit message · Marko Bencun

rust: async unit tests instead of block_on

Easier to read/write the unit tests.

55/100 · ThinMessage clarity
✓ Descriptive subject✓ Names a concrete action or component✓ Mentions testing or verification! No meaningful explanatory body
Why it was queued
cryptography-sensitive pathsigning or wallet pathauthentication path
AI analysis · Informational 15/100

This commit is a code-quality refactor for unit tests only. It introduces a small Rust helper macro called async_test::test that lets developers write async test functions directly, instead of manually wrapping async code in a block_on() call. The change touches many test files but does not alter any production firmware behavior, user-facing functionality, or security logic.

Security candidatemessages: add eth payment request fieldsby Cedric Wiese · 21a01dca · Mar 29, 2026 · 5 filesMessage 45 · ThinInformational 17Details
Commit message · Cedric Wiese

messages: add eth payment request fields

45/100 · ThinMessage clarity
✓ Descriptive subject✓ Names a concrete action or component! No meaningful explanatory body
Why it was queued
cryptography-sensitive pathsigning or wallet path
AI analysis · Informational 17/100

This commit adds a new optional 'payment request' field to Ethereum EIP-1559 signing requests in the BitBox02 firmware's protocol definitions and generated code. It does not contain any actual implementation of how that field is processed, validated, or shown to the user. Based only on the code changed here, there is no identifiable security vulnerability.

Security candidatePort USB report queue to Rustby Niklas Dusenlund · 214254aa · Mar 26, 2026 · 42 filesMessage 78 · AdequateLow 34Details
Commit message · Niklas Dusenlund

Port USB report queue to Rust

Introduce a dedicated bitbox-usb-report-queue crate and expose
an opaque C FFI for allocating, clearing, pushing, peeking and
pulling 64-byte USB reports.

Wire usb_processing, usb_frame, usb_packet and u2f_packet to
use RustUsbReportQueue handles instead of the legacy C queue,
and initialize separate HWW and U2F queues from Rust, the
bootloader, simulators and HID test fakes.

Remove src/queue.c/h and the old bitbox02 queue wrapper, and
update the simulator and test code to drain and free the new
queues.

78/100 · AdequateMessage clarity
✓ Descriptive subject✓ Names a concrete action or component✓ Provides detailed explanatory context✓ Mentions testing or verification
Why it was queued
update trustboot or update path
AI analysis · Low 34/100

This commit rewrites the firmware's USB report queue from C to Rust. It is a large refactoring that replaces a hand-written C ring buffer with a Rust VecDeque wrapped in a C-compatible FFI. The change touches the core USB communication path used by both the main hardware wallet (HWW) and U2F interfaces. There is no explicit security bug fixed or introduced in the diff, but any mistake in the boundary between C and Rust could affect how the device receives and sends USB messages.

Security candidateeth: implement streaming for typed data messagesby Tomas Vrba · fd2653cf · Mar 25, 2026 · 12 filesMessage 78 · AdequateLow 37Details
Commit message · Tomas Vrba

eth: implement streaming for typed data messages

eth: add typed data streaming to python API

show size of truncated string, show ellipsis

add a multiline string test case

- adds a test case to make use of multiline string handling code from in
the .js test generator
- adjust rust test loader to handle strings, not just bytes

78/100 · AdequateMessage clarity
✓ Descriptive subject✓ Names a concrete action or component✓ Provides detailed explanatory context✓ Mentions testing or verification
Why it was queued
cryptography-sensitive pathsigning or wallet path
AI analysis · Low 37/100

This commit extends an existing Ethereum data-streaming feature so it also works for EIP-712 typed messages with large data fields (e.g., a 50 KB 'bytes' field). Previously, only large Ethereum transaction data could be streamed in chunks; now the same chunking is applied to typed-message values. The change also improves on-screen display of long values by warning the user when a value is too large to show in full and by showing an ellipsis. It is a feature addition with defensive input checks, not a fix for a known vulnerability.

Security candidateRevert "simulator: LVGL init"by Niklas · 8f6e8a8a · Mar 23, 2026 · 391 filesMessage 35 · OpaqueInformational 11Details
Commit message · Niklas

Revert "simulator: LVGL init"

35/100 · OpaqueMessage clarity
✓ Descriptive subject! No meaningful explanatory body! Opaque security-relevant change
Why it was queued
seed or entropy pathauthentication pathparser or protocol path
AI analysis · Informational 11/100

This commit is a large revert that removes a previously added graphical simulator feature (LVGL init) from the BitBox02 firmware repository. It deletes the LVGL submodule, vendored Rust dependencies, simulator code, and related build files. There is no indication in the commit message or diff that this change fixes a security vulnerability; it appears to be a routine rollback of an unfinished or unwanted feature.

Security candidatesimulator: LVGL initby Niklas Dusenlund · 00a6e4d7 · Mar 23, 2026 · 391 filesMessage 28 · OpaqueInformational 15Details
Commit message · Niklas Dusenlund

simulator: LVGL init

28/100 · OpaqueMessage clarity
✓ Subject identifies a change! No meaningful explanatory body! Opaque security-relevant change
Why it was queued
seed or entropy pathauthentication pathparser or protocol path
AI analysis · Informational 15/100

This commit adds a new graphical simulator for the BitBox02 hardware wallet using the LVGL UI library. It is a large development-only change that introduces simulator code, vendored Rust dependencies, and build tooling. There is no indication from the commit title, message, or diff that this fixes or introduces any security vulnerability.

Security candidateAdd coin purchase memo supportby cedwies · 10746100 · Feb 26, 2026 · 8 filesMessage 45 · ThinLow 27Details
Commit message · cedwies

Add coin purchase memo support

45/100 · ThinMessage clarity
✓ Descriptive subject✓ Names a concrete action or component! No meaningful explanatory body
Why it was queued
cryptography-sensitive pathsigning or wallet path
AI analysis · Low 27/100

This commit adds a new 'coin purchase memo' feature to BitBox02 payment requests. It lets a payment request include details about a separate coin purchase (e.g., '0.25 ETH to address 0x...') and verifies that the listed Ethereum address really belongs to the wallet by deriving it from a provided keypath. The change is mostly a feature addition, but it introduces cross-currency validation logic and a TODO noting that the user-interface confirmation for this new memo type is not yet implemented.

Security candidatebitbox02-rust: remove redundant mock_memory callsby Marko Bencun · c7089da8 · Feb 19, 2026 · 11 filesMessage 60 · AdequateInformational 15Details
Commit message · Marko Bencun

bitbox02-rust: remove redundant mock_memory calls

With introduction of HAL, many of them are not needed anymore.

60/100 · AdequateMessage clarity
✓ Descriptive subject✓ Names a concrete action or component✓ Provides an explanatory body
Why it was queued
signing or wallet pathauthentication path
AI analysis · Informational 15/100

This commit only cleans up test code by removing unnecessary calls to a mock memory setup function. It does not change the actual firmware that runs on the BitBox02 device, so it has no security impact on users.

Security candidatekeystore: add KeystoreHal abstractionby Marko Bencun · 4fc335ee · Feb 19, 2026 · 2 filesMessage 58 · ThinInformational 15Details
Commit message · Marko Bencun

keystore: add KeystoreHal abstraction

Introduce the KeystoreHal trait and KeystoreHalImpl adapter over HAL
subsystems.

Route unlock_bip39 and transitive keystore internals through
KeystoreHal and update workflow unlock to pass KeystoreHalImpl.

Reason: to not keep a mutable borrow of Hal Ui (part of Hal), which
will be used in parallel during unlock_bip39(), which performs an UI
animation in parallel.

58/100 · ThinMessage clarity
✓ Descriptive subject✓ Provides detailed explanatory context
Why it was queued
secret or key material
AI analysis · Informational 15/100

This commit is a straightforward internal code refactor. It introduces a new 'KeystoreHal' adapter so that the keystore code can borrow only the hardware subsystems it needs (memory, random number generator, secure chip) instead of borrowing the entire HAL, which also contains the user interface. The stated reason is to allow a UI animation to run in parallel while the device unlocks. There is no change to security logic, cryptography, or user-facing behavior.

Security candidatesalt: hash_data takes Memoryby Marko Bencun · b608856e · Feb 19, 2026 · 3 filesMessage 58 · ThinInformational 12Details
Commit message · Marko Bencun

salt: hash_data takes Memory

Change salt::hash_data to take &mut impl Memory instead of &mut impl Hal.

Update keystore callsites to pass hal.memory().

Reason: to not keep a mutable borrow of Hal Ui (part of Hal), which
will be used in parallel during unlock_bip39().

58/100 · ThinMessage clarity
✓ Descriptive subject✓ Provides detailed explanatory context
Why it was queued
secret or key materialboot or update path
AI analysis · Informational 12/100

This is a small internal code cleanup in the BitBox02 firmware's Rust code. It changes a helper function so it borrows only the memory/storage part of the hardware abstraction layer, rather than the whole hardware abstraction layer. The commit message says this is preparation for a future change where the user interface part of the hardware layer will be used at the same time during wallet unlocking. There is no direct security fix here and no vulnerability is described.

Security candidatebitbox02_rust/hal: add ui confirm/fontby Marko Bencun · 5bacbf78 · Feb 18, 2026 · 38 filesMessage 78 · AdequateInformational 15Details
Commit message · Marko Bencun

bitbox02_rust/hal: add ui confirm/font

Introduce hal::ui::ConfirmParams and hal::ui::Font so HAL interfaces no longer depend on workflow or bitbox02 ui confirmation types.

Adapt all callsites and add exhaustive conversion tests in hal::bitbox02::ui for font and confirm params mapping.

78/100 · AdequateMessage clarity
✓ Descriptive subject✓ Names a concrete action or component✓ Provides detailed explanatory context✓ Mentions testing or verification
Why it was queued
signing or wallet pathauthentication path
AI analysis · Informational 15/100

This commit is a routine internal code cleanup. It moves the definitions of confirmation-screen parameters and fonts out of the workflow layer and into the hardware-abstraction (HAL) layer, then updates all call sites to use the new names. There is no change to user-facing behavior, no bug fix, and no security vulnerability being addressed.

Security candidatedocker: Add new targetby Niklas Dusenlund · 57410cb2 · Feb 16, 2026 · 423 filesMessage 63 · AdequateInformational 17Details
Commit message · Niklas Dusenlund

docker: Add new target

* Add target for cortex-m33
* Update rustc to latest (reduces firmware size by about 4k)
* Update llvm to match rust
* Vendor new deps due to new toolchain
* Update Arm GNU toolchain to 15 (further reduces by about 2k)

63/100 · AdequateMessage clarity
✓ Subject identifies a change✓ Provides detailed explanatory context✓ Explains rationale or failure mode
Why it was queued
cryptography-sensitive pathseed or entropy pathsigning or wallet pathauthentication pathparser or protocol path
AI analysis · Informational 17/100

This is a routine build-system and toolchain update for the BitBox02 firmware. It adds support for a new ARM processor target (Cortex-M33), upgrades the Rust compiler, LLVM, and the ARM GNU toolchain, and pulls in newer versions of some third-party Rust libraries. There is no direct evidence in the commit of a security vulnerability being fixed or introduced. The main risk is the usual one with large toolchain and dependency updates: a small chance that compiler, library, or behavior changes could introduce subtle bugs, but nothing in the diff points to an exploitable flaw.

Security candidateeth: allow EIP-712 message signing without anti-kleptoby benma's agent · 24ca04e8 · Feb 15, 2026 · 11 filesMessage 78 · AdequateLow 31Details
Commit message · benma's agent

eth: allow EIP-712 message signing without anti-klepto

Some defi apps require deterministic signatures, and anti-klepto is
not compatible with, as a random host nonce is contributed.

The host could instead contribute a constant nonce, which would work,
but the resulting signature would not be the same as a regular RFC6979
signature that does not use the additional data. This would not be
optimal in terms of interopability with other wallets.

This commit allows the host to skip the host nonce, in which case we
fall back to regular deterministic signatures.

78/100 · AdequateMessage clarity
✓ Specific, descriptive subject✓ Names a concrete action or component✓ Provides detailed explanatory context✓ Names security-relevant behavior explicitly
Why it was queued
signing boundarysecure hardware boundarycryptography-sensitive pathsigning or wallet path
AI analysis · Low 31/100

This commit adds an optional mode for signing Ethereum typed messages (EIP-712) that skips the anti-klepto protocol, falling back to normal deterministic signatures. The change is intentional and documented: some DeFi apps need signatures that are reproducible by other wallets, which the anti-klepto protocol prevents. The feature is gated behind a new host flag and requires firmware v9.26.0 or newer. It does not remove anti-klepto from other signing paths such as Bitcoin or regular Ethereum transactions.

Security candidateworkflow/mnemonic: use HAL to get random numberby Marko Bencun · dcb05fbc · Feb 14, 2026 · 6 filesMessage 50 · ThinLow 29Details
Commit message · Marko Bencun

workflow/mnemonic: use HAL to get random number

50/100 · ThinMessage clarity
✓ Descriptive subject✓ Names a concrete action or component✓ Names security-relevant behavior explicitly! No meaningful explanatory body
Why it was queued
entropy or randomness
AI analysis · Low 29/100

This commit refactors how random numbers are generated during the backup verification workflow. Previously, the code called a low-level device function directly; now it uses a hardware abstraction layer (HAL) interface so the same code can be tested with a mock random source. The change itself is a code-quality/testability improvement and does not appear to fix an active security vulnerability, but it touches the code that creates the random word-order challenges used to verify a user's seed backup.

Security candidateeth: add tests with large data to test streamingby Tomas Vrba · 2a1108e0 · Feb 13, 2026 · 6 filesMessage 78 · AdequateInformational 15Details
Commit message · Tomas Vrba

eth: add tests with large data to test streaming

Adds transactions with large data (about 20% of total txs) to the
sighash go script to unit test the chunking producer

- test sign.rs to test that new streaming mode produces same
signatures as non-streaming
- test that streaming large data produces signatures of expected size

eth: prepare tests for eth sighash streaming

prepares initial unit tests for simple and chunking producer for data
streaming

tests generated by Claude Code

78/100 · AdequateMessage clarity
✓ Descriptive subject✓ Names a concrete action or component✓ Provides detailed explanatory context✓ Mentions testing or verification
Why it was queued
signing boundarysigning or wallet path
AI analysis · Informational 15/100

This commit is purely a test-suite update for the BitBox02 hardware wallet's Ethereum transaction signing. It adds unit tests that verify the new 'streaming' mode for large transaction data produces the same cryptographic signatures as the older non-streaming mode, and updates test data files. There is no change to production firmware code or user-facing behavior, and nothing in the commit suggests a security vulnerability.

Security candidateeth: prepare and generate proto files for data streamingby Tomas Vrba · 83be97ad · Feb 13, 2026 · 6 filesMessage 83 · StrongInformational 11Details
Commit message · Tomas Vrba

eth: prepare and generate proto files for data streaming

update existing tests to include the new data length field

Adds new protobuf fields to support streaming large transaction data:
- data_length: Indicates total data size (triggers streaming mode if > 6144)
- EthSignDataRequestChunkResponse: Request for a data chunk at offset/length
- EthSignDataResponseChunkRequest: Response containing the requested chunk

Regenerates all protobuf files (Python, Rust, C) from the updated .proto.

Updates existing tests to populate the new data_length field (set to 0
for traditional mode where all data is sent inline).

83/100 · StrongMessage clarity
✓ Specific, descriptive subject✓ Names a concrete action or component✓ Provides detailed explanatory context✓ Mentions testing or verification
Why it was queued
cryptography-sensitive pathsigning or wallet path
AI analysis · Informational 11/100

This commit adds new message definitions and generated code to support streaming large Ethereum transaction data in chunks, but it does not implement the actual streaming logic. Existing tests are updated only to include the new optional field set to zero, meaning the current behavior is unchanged. There is no indication this is a security fix or introduces a vulnerability.

Security candidatehal: remove the crate::workflow::testing re-exportby benma's agent · 9988cb33 · Feb 10, 2026 · 22 filesMessage 60 · AdequateInformational 15Details
Commit message · benma's agent

hal: remove the crate::workflow::testing re-export

60/100 · AdequateMessage clarity
✓ Specific, descriptive subject✓ Names a concrete action or component✓ Mentions testing or verification! No meaningful explanatory body
Why it was queued
signing or wallet pathauthentication path
AI analysis · Informational 15/100

This commit is a simple internal cleanup in the BitBox02 firmware's Rust code. It removes a shortcut (a 're-export') that made test-only user-interface helpers available under an old module path, and updates all test files to import those helpers directly from their new location. There is no change to the actual device behavior, no bug fix, and no security impact.

Security candidatebuild: enable -Wcast-align=strictby benma's agent · ba2124e5 · Feb 3, 2026 · 11 filesMessage 98 · StrongModerate 53Details
Commit message · benma's agent

build: enable -Wcast-align=strict

Enable GCC -Wcast-align=strict globally (Clang uses -Wcast-align) so
misaligned pointer-cast issues are caught across firmware, bootloader,
unit tests, and simulators.

Fixes fall into two UB classes:

1) Alignment UB: casting byte buffers/packed payloads to wider pointer
types and then dereferencing can require stricter alignment than the
source object provides. This is undefined behavior per C.

2) Effective-type / strict-aliasing UB: reading a value by
reinterpreting a uint8_t buffer as a different object type via a
pointer cast violates C’s effective-type/aliasing rules, so the
compiler may miscompile even when the address happens to be aligned.

Resolve by memcpy’ing into
properly-typed locals (e.g. USB_FRAME, u32/version_t) before use.

Useful references:
- https://www.open-std.org/jtc1/sc22/WG14/www/docs/n3519.pdf

For C11 (ISO/IEC 9899:2011), the key places are:

- Alignment UB from pointer casts: §6.3.2.3 p7 (“Pointers”) — converting to a different object pointer type and the result not being correctly aligned is UB.
- Effective type rule: §6.5 p6 (“Expressions”) — defines an object’s effective type (including the memcpy/memmove wording).
- Strict-aliasing rule: §6.5 p7 — lists the allowed lvalue types you may use to access an object’s stored value (the classic aliasing bullet list).

98/100 · StrongMessage clarity
✓ Descriptive subject✓ Names a concrete action or component✓ Uses a recognizable type or scope✓ Provides detailed explanatory context✓ Mentions testing or verification✓ Links an issue, advisory, or supporting reference
Why it was queued
update trustboot or update path
AI analysis · Moderate 53/100

This commit turns on a stricter compiler warning that catches risky pointer casts in C code, and fixes the resulting warnings. The changes replace direct casts from byte buffers to larger types (like treating a raw byte array as a 32-bit number or USB frame) with safer memcpy-based copies. The commit message explicitly frames these as classes of undefined behavior (alignment and strict-aliasing issues) that could lead to miscompilation or incorrect reads. It is a hardening/correctness patch rather than a fix for a known exploitable bug, but the affected code paths include bootloader pairing-code handling, firmware version parsing, USB packet processing, and BLE pairing-code display.

Security candidatebitbox-secp256k1: remove needless ffi def of secp256k1_ecdsa_signature_serialize_compactby benma's agent · e338d122 · Feb 3, 2026 · 1 fileMessage 65 · AdequateInformational 15Details
Commit message · benma's agent

bitbox-secp256k1: remove needless ffi def of secp256k1_ecdsa_signature_serialize_compact

It is already wrapped in the bitcoin::secp256k1 dep.

65/100 · AdequateMessage clarity
✓ Specific, descriptive subject✓ Names a concrete action or component✓ Provides an explanatory body
Why it was queued
signing boundarycryptography-sensitive path
AI analysis · Informational 15/100

This commit removes a duplicate low-level binding to a secp256k1 cryptographic function and replaces its use with an equivalent, already-vetted wrapper from the project's dependency on the bitcoin::secp256k1 Rust crate. It is a code cleanup/refactoring change with no security-relevant behavior change visible in the diff.

Security candidatebuild: consolidate secp256k1 into new crate bitbox-secp256k1by benma's agent · eabe5aa4 · Feb 3, 2026 · 27 filesMessage 85 · StrongInformational 12Details
Commit message · benma's agent

build: consolidate secp256k1 into new crate bitbox-secp256k1

- cargo-first build of the secp256k1-zkp C dep, so the crate can be
compiled without CMake and normal cargo tooling works
- Moved bitbox02::secp256k1 and bitbox02_rust::secp256k1 into it

85/100 · StrongMessage clarity
✓ Specific, descriptive subject✓ Names a concrete action or component✓ Uses a recognizable type or scope✓ Provides detailed explanatory context
Why it was queued
cryptography-sensitive pathsigning or wallet path
AI analysis · Informational 12/100

This commit is a build-system refactor: it moves the secp256k1 cryptographic library dependency into a new Rust crate called bitbox-secp256k1 so it can be built with normal Cargo tooling instead of CMake. The actual cryptographic functions (signing, nonce commitments, DLEQ proofs, etc.) are copied almost unchanged from existing modules into the new crate. There is no direct evidence in the commit that this fixes a security vulnerability or introduces a new exploit.

Security candidatefactorysetup: move secp256k1 verification to Rustby benma's agent · e27a884e · Feb 3, 2026 · 4 filesMessage 78 · AdequateInformational 20Details
Commit message · benma's agent

factorysetup: move secp256k1 verification to Rust

The last remaining secp256k1 C library use. After this, we can
consolidate all sec256k1 things into a new bitbox-secp256k1 crate.

78/100 · AdequateMessage clarity
✓ Descriptive subject✓ Names a concrete action or component✓ Provides detailed explanatory context✓ Mentions testing or verification
Why it was queued
defensive validationcryptography-sensitive path
AI analysis · Informational 20/100

This commit rewrites a small piece of the BitBox02 factory-setup code from C to Rust. The code verifies a cryptographic signature on the device during factory setup. The change itself is a routine refactoring: it removes the old C secp256k1 library call and replaces it with an equivalent Rust call. There is no indication in the commit that this fixes a known security bug, and the new Rust code includes tests for invalid inputs. On its own, this looks like a defensive hardening/cleanup change rather than a vulnerability fix.

Security candidateport communication_mode to Rustby benma's agent · bc518af8 · Jan 16, 2026 · 15 filesMessage 58 · ThinInformational 17Details
Commit message · benma's agent

port communication_mode to Rust

With extern C functions to use them in the C code. The firmware main
loop will be ported to Rust, so it will not be needed there much
longer, but the bootloader also uses it.

58/100 · ThinMessage clarity
✓ Descriptive subject✓ Provides detailed explanatory context
Why it was queued
update trustboot or update pathparser or protocol path
AI analysis · Informational 17/100

This commit rewrites a small piece of firmware logic from C to Rust. The logic decides whether the BitBox02 Nova hardware wallet should use Bluetooth or USB for communication. There is no obvious security bug introduced by the change; it appears to be a routine language port with matching behavior and added unit tests.

Security candidaterust: update rust-miniscript to v13.0.0.by Marko Bencun · 4c1c8000 · Jan 14, 2026 · 122 filesMessage 60 · AdequateInformational 16Details
Commit message · Marko Bencun

rust: update rust-miniscript to v13.0.0.

This saves 30144 bytes in the firmware binary.

60/100 · AdequateMessage clarity
✓ Descriptive subject✓ Names a concrete action or component✓ Provides an explanatory body
Why it was queued
signing or wallet pathboot or update pathparser or protocol path
AI analysis · Informational 16/100

This commit updates the BitBox02 firmware's vendored copy of the Rust 'miniscript' library to version 13.0.0, along with its dependency 'hex-conservative'. The stated reason is to reduce firmware size by about 30 KB. The change is a routine dependency refresh; the commit message and diff do not describe any security bug being fixed, and no independent security advisory is supplied.