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.
Message quality measures whether a commit identifies its scope, purpose, rationale, testing, and supporting references. It does not change the security-severity score.
This commit fixes a small but meaningful bug in the Ledger Bitcoin app's code that handles advanced multi-signature (MuSig) operations. A function called crypto_tr_lift_x can fail when given an invalid x-coordinate that does not correspond…
Unchecked return value from cryptographic point-lifting functionPotential use of invalid/uninitialized curve point in MuSig signing flowMissing error propagation in compressed public-key parsing
This commit hardens a Bitcoin transaction parser in Ledger's app. It changes some numeric fields to safer types, adds bounds checks on transaction counts and sizes, rejects extra trailing bytes after a transaction, checks a previously-igno…
Integer type narrowing and sentinel change from -1 to SIZE_MAXBounds checks before uint64_t to unsigned int castsReturn value of parser_consolidate_buffers now checked
This commit changes what information the Ledger Bitcoin app shows on its screen when signing a transaction that includes 'external inputs'—coins the user did not provide and whose amounts the device cannot independently verify. Previously …
UI/UX hardening: additional contextual data shown for transactions with unverified external inputsNo change to cryptographic checks, sighash validation, or input verificationExternal inputs warning is preserved; new data is displayed only when the input set is closed (no ANYONECANPAY)
This commit only updates test screenshot snapshots for the Ledger Bitcoin app. It does not change any application source code, so it cannot directly fix or introduce a security vulnerability in the shipped app. The snapshots reflect new ex…
No source-code changes in the commitOnly test snapshot images (PNG) are modified or addedCommit title references trustworthy amount/fee display for non-default sighash, suggesting a prior functional change exists elsewhere
This commit only adds and updates automated integration tests for the Ledger Bitcoin app. It does not change the app's actual signing code. The tests verify that the device now shows clearer, more trustworthy on-screen amount and fee infor…
No C/device firmware code changesTest-only commit adjusting UI navigation expectationsNew tests assert that negative fees are rejected for default sighash but allowed/relabeled for ANYONECANPAY sighash
This commit only adds new automated unit tests for functions that decide how transaction amounts and fees are displayed on a Ledger device when signing a Bitcoin transaction with non-standard signature hash (sighash) modes. It does not cha…
Commit title references trustworthy amount/fee display for non-default sighash, a historically sensitive area in hardware-wallet signing UITests cover sighash commitment semantics (SIGHASH_NONE, SIGHASH_SINGLE, SIGHASH_ANYONECANPAY combinations) that directly affect what outputs a signature commits toNo functional code changes in this commit; only unit-test additions
This commit changes how the Ledger Bitcoin app shows transaction amounts and fees on the device screen when a user signs a transaction that does not use the normal 'sign everything' rule (a so-called non-default sighash). Previously the ap…
This commit is a small code cleanup in the Ledger Bitcoin app. It removes an initial loop that counted commands and instead checks for the end marker directly while processing. The author says the old loop was unnecessary and caused a harm…
Commit message explicitly frames the change as removing a redundant scan and a static-analysis false positive, not as a security fix.No change to parsing, cryptography, authorization, or transaction semantics.Functional behavior for valid command lists is equivalent before and after the patch.
This commit only fixes a documentation/declaration inconsistency in a C header file. The function parameter name in the comment is changed from `in_len` to `inlen` to match the actual declaration, and the output pointer is annotated with a…
This commit adds test-only helper code for creating Bitcoin transaction fixtures that support newer wallet types (MuSig2 multi-signature and several legacy/wrapped SegWit descriptor forms). It is not a change to the Ledger app firmware its…
No security-relevant signal: change is in test utilities onlyAdds cryptographic helper for MuSig2 key aggregation in test fixturesAdds P2SH/P2SH-P2WPKH/P2SH-P2WSH script generation in test fixtures
This commit is purely a build-system and test-harness cleanup. It refactors CMakeLists.txt, updates comments, simplifies the speculos test bridge, and removes a workaround in a test file. Nothing here changes the actual Ledger Bitcoin app …
This commit only adds new unit tests and renames some existing test functions. It does not change any production code behavior, so it cannot introduce a security vulnerability or fix one on its own.
This commit only adds unit tests and a test-only bridge implementation for cryptographic signing functions. It does not change the actual device application code that users rely on, so it cannot introduce a security vulnerability in the sh…
This commit is a test-only refactoring: it removes a set of local mock SDK headers and a software SHA-256 implementation used by unit tests, and instead makes those tests run against the real Ledger SDK headers through the existing Speculo…
No production code changesTest infrastructure refactoring onlyRemoval of local cryptographic mock in favor of real SDK/speculos bridge
This commit is purely about improving the project's automated test setup. It links unit tests to a software simulator (Speculos) so cryptographic code can be tested on a regular computer without a physical Ledger device. It adds tests for …
This commit is purely a testing infrastructure cleanup. It removes a workaround called SKIP_FOR_CMOCKA that previously hid parts of the code from unit tests, and instead compiles those parts using a more realistic emulator bridge (speculos…
This commit only adds unit tests and test-support code. It does not change the actual Bitcoin app that runs on Ledger devices, so it cannot introduce a security vulnerability in shipped firmware. The changes make the test harness able to s…
This commit only adds helper functions for unit tests to convert Bitcoin extended public keys (xpubs) between base58 string form and raw binary form. It does not change the actual Ledger app code that runs on the device, nor does it fix or…
This commit only adds unit tests and test infrastructure for Bitcoin Taproot cryptographic helpers. It does not change the actual app code that runs on the hardware wallet, so it cannot introduce a security vulnerability in shipped firmwar…
This commit is a routine cleanup of the project's unit-test code. It removes duplicate mock implementations of a small helper function called `pic()` from 14 test files and places a single shared version in a mock header. The change only a…
✓ Descriptive subject✓ Names a concrete action or component! No meaningful explanatory body
Why it was queued
cryptography-sensitive path
AI analysis · Moderate 59/100
This commit fixes a small but meaningful bug in the Ledger Bitcoin app's code that handles advanced multi-signature (MuSig) operations. A function called crypto_tr_lift_x can fail when given an invalid x-coordinate that does not correspond to any real point on the Bitcoin elliptic curve. Previously, the cpoint function ignored that failure and kept using the resulting output as if it were valid. The patch now checks the return value, prints an error message, and returns an error code instead of continuing with potentially bad data. In a hardware wallet, using invalid curve points could in theory lead to incorrect signature calculations or unexpected behavior, though the practical exploit path is not fully clear from the diff alone.
Security candidateType consistency in psbt_parse_rawtx, and some other hardeningsby Salvatore Ingala · 1af86279 · Jul 31, 2026 · 3 filesMessage 73 · AdequateModerate 59Details
Commit message · Salvatore Ingala
Type consistency in psbt_parse_rawtx, and some other hardenings
Change key_len and output_index to size_t for consistency with other APIs, add some extra bounds checks.
Also adds: - missing return value check on parser_consolidate_buffers - reject trailing data during parsing - zero txid_parser_outputs_t before starting parsing
73/100 · AdequateMessage clarity
✓ Specific, descriptive subject✓ Names a concrete action or component✓ Provides detailed explanatory context
Why it was queued
signing boundarydefensive validationsigning or wallet path
AI analysis · Moderate 59/100
This commit hardens a Bitcoin transaction parser in Ledger's app. It changes some numeric fields to safer types, adds bounds checks on transaction counts and sizes, rejects extra trailing bytes after a transaction, checks a previously-ignored internal buffer operation, and zeroes out result memory before use. These are defensive fixes that could prevent memory corruption, incorrect parsing, or information leakage, but the commit does not describe a specific active exploit.
Document signing behavior with sighash flags, and external inputs
50/100 · ThinMessage clarity
✓ Specific, descriptive subject✓ Names a concrete action or component! No meaningful explanatory body
Why it was queued
signing boundarydocumentation-only discount
Security candidateDisplay total amount of external inputs when reliableby Salvatore Ingala · 26f631c5 · Jul 20, 2026 · 4 filesMessage 78 · AdequateInformational 21Details
Commit message · Salvatore Ingala
Display total amount of external inputs when reliable
While warning for the presence of external inputs remains necessary (certain attacks are possible), we can show additional info when the amount of external inputs is committed to (that is, when the sighash flag we're signing with does not have ANYONECANPAY, and therefore the signature commits to all inputs).
In that case, we can show: - The total amount of external inputs - The net spend (or receive) from/to the wallet policy, similarly to the transaction summary shown with certain sighash flags.
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
explicit security languagesigning boundarysigning or wallet path
AI analysis · Informational 21/100
This commit changes what information the Ledger Bitcoin app shows on its screen when signing a transaction that includes 'external inputs'—coins the user did not provide and whose amounts the device cannot independently verify. Previously the app only warned that such inputs exist. After this change, when the signature commits to the full set of inputs (i.e., no ANYONECANPAY sighash flag), the app also displays the total value of those external inputs and the net amount the user's wallet is spending or receiving. This is a user-interface improvement intended to reduce confusion, not a fix for a code-execution or theft vulnerability. The underlying security model—external inputs are still 'unverified'—is unchanged.
Trustworthy sign_psbt amount/fee display for non-default sighash: snapshots
50/100 · ThinMessage clarity
✓ Specific, descriptive subject✓ Names a concrete action or component! No meaningful explanatory body
Why it was queued
signing boundarysigning or wallet path
AI analysis · Informational 15/100
This commit only updates test screenshot snapshots for the Ledger Bitcoin app. It does not change any application source code, so it cannot directly fix or introduce a security vulnerability in the shipped app. The snapshots reflect new expected screen output after a prior code change that altered how amounts and fees are displayed when signing transactions with non-default Bitcoin sighash types.
Trustworthy sign_psbt amount/fee display for non-default sighash: integration tests
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 boundarysigning or wallet path
AI analysis · Informational 12/100
This commit only adds and updates automated integration tests for the Ledger Bitcoin app. It does not change the app's actual signing code. The tests verify that the device now shows clearer, more trustworthy on-screen amount and fee information when a transaction uses unusual Bitcoin signature modes (non-default 'sighash' types). Because no production firmware code is modified, there is no direct security vulnerability or fix introduced by this commit itself.
Trustworthy sign_psbt amount/fee display for non-default sighash: unit-tests
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 boundary
AI analysis · Informational 12/100
This commit only adds new automated unit tests for functions that decide how transaction amounts and fees are displayed on a Ledger device when signing a Bitcoin transaction with non-standard signature hash (sighash) modes. It does not change the actual signing or display logic, so by itself it cannot introduce or fix a security bug. The tests appear to be a companion to a separate code change that made the amount/fee display trustworthy for non-default sighash types.
Trustworthy sign_psbt amount/fee display for non-default sighash: implementation
50/100 · ThinMessage clarity
✓ Specific, descriptive subject✓ Names a concrete action or component! No meaningful explanatory body
Why it was queued
signing boundarysigning or wallet path
AI analysis · Moderate 59/100
This commit changes how the Ledger Bitcoin app shows transaction amounts and fees on the device screen when a user signs a transaction that does not use the normal 'sign everything' rule (a so-called non-default sighash). Previously the app might have displayed a fee or total that could actually change later, because some sighash types let other people add inputs or change outputs after the signature. Now the app detects those cases and either shows only the net amount the user is spending/receiving, or shows a warning that amounts cannot be verified. It also adds a 'Signing rule' line so the user knows the transaction is not fully committed.
Security candidateFix slight mismatch in definition of crypto_hash160by Salvatore Ingala · 54587441 · Jun 3, 2026 · 1 fileMessage 50 · ThinInformational 15Details
Commit message · Salvatore Ingala
Fix slight mismatch in definition of crypto_hash160
50/100 · ThinMessage clarity
✓ Specific, descriptive subject✓ Names a concrete action or component! No meaningful explanatory body
Why it was queued
cryptography-sensitive path
AI analysis · Informational 15/100
This commit only fixes a documentation/declaration inconsistency in a C header file. The function parameter name in the comment is changed from `in_len` to `inlen` to match the actual declaration, and the output pointer is annotated with a fixed-size array hint (`out[static 20]`) instead of a plain pointer. There is no change to executable code or behavior.
It is redundant, it's simpler to just exit when the sentinel is found. It was also making clang static analyzer detect a false positive out of bounds read.
68/100 · AdequateMessage clarity
✓ Descriptive subject✓ Names a concrete action or component✓ Provides detailed explanatory context
Why it was queued
memory safety
AI analysis · Informational 11/100
This commit is a small code cleanup in the Ledger Bitcoin app. It removes an initial loop that counted commands and instead checks for the end marker directly while processing. The author says the old loop was unnecessary and caused a harmless false-positive warning in a static analysis tool. There is no indication this fixes a real security bug or changes app behavior in a user-visible way.
Security candidateAdd support for generating PSBTs for MuSig2 in txmakerby Salvatore Ingala · 13d84aca · Jun 1, 2026 · 2 filesMessage 65 · AdequateInformational 19Details
Commit message · Salvatore Ingala
Add support for generating PSBTs for MuSig2 in txmaker
Also, address the remaining some old TODOs.
65/100 · AdequateMessage clarity
✓ Specific, descriptive subject✓ Names a concrete action or component✓ Provides an explanatory body
Why it was queued
signing boundarysigning or wallet path
AI analysis · Informational 19/100
This commit adds test-only helper code for creating Bitcoin transaction fixtures that support newer wallet types (MuSig2 multi-signature and several legacy/wrapped SegWit descriptor forms). It is not a change to the Ledger app firmware itself, but to Python utilities used in automated tests. There is no obvious security bug introduced, and the commit does not claim to fix a vulnerability.
Security candidateAdd unit tests for crypto_ripemd160, crypto_hash160, crypto_get_checksumby Salvatore Ingala · af900978 · May 29, 2026 · 2 filesMessage 60 · AdequateInformational 15Details
Commit message · Salvatore Ingala
Add unit tests for crypto_ripemd160, crypto_hash160, crypto_get_checksum
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
cryptography-sensitive path
AI analysis · Informational 15/100
This commit only adds new unit tests and renames some existing test functions. It does not change any production code behavior, so it cannot introduce a security vulnerability or fix one on its own.
Security candidateDelete mock_include, rely on the speculos_bridge for all testsby Salvatore Ingala · 6f7b4472 · May 29, 2026 · 28 filesMessage 60 · AdequateInformational 15Details
Commit message · Salvatore Ingala
Delete mock_include, rely on the speculos_bridge for all tests
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
cryptography-sensitive path
AI analysis · Informational 15/100
This commit is a test-only refactoring: it removes a set of local mock SDK headers and a software SHA-256 implementation used by unit tests, and instead makes those tests run against the real Ledger SDK headers through the existing Speculos emulator bridge. No production application code is changed, and there is no indication this fixes or introduces a security vulnerability.
Security candidateAdd unit tests for crypto_get_compressed_pubkey_at_pathby Salvatore Ingala · 86fa0b30 · May 29, 2026 · 4 filesMessage 60 · AdequateInformational 15Details
Commit message · Salvatore Ingala
Add unit tests for crypto_get_compressed_pubkey_at_path
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
cryptography-sensitive path
AI analysis · Informational 15/100
This commit only adds unit tests and test-support code. It does not change the actual Bitcoin app that runs on Ledger devices, so it cannot introduce a security vulnerability in shipped firmware. The changes make the test harness able to simulate BIP32 key derivation using the Speculos emulator's own implementation, with a fixed public test seed.
Security candidateAdd minimal utility functions to work with serialized xpub in unit testsby Salvatore Ingala · 31a302d2 · May 29, 2026 · 4 filesMessage 75 · AdequateInformational 15Details
Commit message · Salvatore Ingala
Add minimal utility functions to work with serialized xpub in unit tests
It is much cleaner than encoding in hex.
75/100 · AdequateMessage clarity
✓ Specific, descriptive subject✓ Names a concrete action or component✓ Provides an explanatory body✓ Mentions testing or verification
Why it was queued
cryptography-sensitive path
AI analysis · Informational 15/100
This commit only adds helper functions for unit tests to convert Bitcoin extended public keys (xpubs) between base58 string form and raw binary form. It does not change the actual Ledger app code that runs on the device, nor does it fix or introduce any security-relevant behavior. It is a test-code cleanup that replaces hard-coded hex byte arrays with readable xpub strings.
Security candidateLink unit tests with speculos implementations of crypto syscallsby Salvatore Ingala · 53a997c6 · May 29, 2026 · 6 filesMessage 75 · AdequateInformational 15Details
Commit message · Salvatore Ingala
Link unit tests with speculos implementations of crypto syscalls
This allows writing unit tests for code that calls syscalls. In this commit, tests for bip32_CKDpub are added.
75/100 · AdequateMessage clarity
✓ Specific, descriptive subject✓ Names a concrete action or component✓ Provides an explanatory body✓ Mentions testing or verification
Why it was queued
secret or key materialcryptography-sensitive path
AI analysis · Informational 15/100
This commit is purely about improving the project's automated test setup. It links unit tests to a software simulator (Speculos) so cryptographic code can be tested on a regular computer without a physical Ledger device. It adds tests for a Bitcoin key-derivation function and re-enables an existing test file. There is no change to the actual app code that runs on the device, and nothing in the commit suggests a security vulnerability or fix.
Security candidateAdd unit tests for crypto_ecdsa_sign_sha256_hash_with_keyby Salvatore Ingala · 96a9203d · May 29, 2026 · 2 filesMessage 60 · AdequateInformational 12Details
Commit message · Salvatore Ingala
Add unit tests for crypto_ecdsa_sign_sha256_hash_with_key
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
cryptography-sensitive path
AI analysis · Informational 12/100
This commit only adds unit tests and a test-only bridge implementation for cryptographic signing functions. It does not change the actual device application code that users rely on, so it cannot introduce a security vulnerability in the shipped product. The changes make the test suite more complete by exercising an existing signing function with a known, precomputed signature.
Security candidateUnit tests for the crypto_tr_* taproot helpersby Salvatore Ingala · cc9ea60d · May 29, 2026 · 2 filesMessage 55 · ThinInformational 15Details
Commit message · Salvatore Ingala
Unit tests for the crypto_tr_* taproot helpers
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 path
AI analysis · Informational 15/100
This commit only adds unit tests and test infrastructure for Bitcoin Taproot cryptographic helpers. It does not change the actual app code that runs on the hardware wallet, so it cannot introduce a security vulnerability in shipped firmware. The changes make the test mock more complete by wiring it to the Speculos emulator's real cryptographic functions instead of crashing with a stub.
Security candidateGet rid of SKIP_FOR_CMOCKA; compile tests with the speculos bridgeby Salvatore Ingala · c098c659 · May 29, 2026 · 10 filesMessage 60 · AdequateInformational 15Details
Commit message · Salvatore Ingala
Get rid of SKIP_FOR_CMOCKA; compile tests with the speculos bridge
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
cryptography-sensitive pathsigning or wallet path
AI analysis · Informational 15/100
This commit is purely a testing infrastructure cleanup. It removes a workaround called SKIP_FOR_CMOCKA that previously hid parts of the code from unit tests, and instead compiles those parts using a more realistic emulator bridge (speculos). No user-facing behavior of the Bitcoin app changes, and no security bug is fixed or introduced in the diff.
Security candidateVarious improvements, refactorings and simplifications from PR reviewby Salvatore Ingala · 1e4853f1 · May 29, 2026 · 4 filesMessage 50 · ThinInformational 15Details
Commit message · Salvatore Ingala
Various improvements, refactorings and simplifications from PR review
50/100 · ThinMessage clarity
✓ Specific, descriptive subject✓ Names a concrete action or component! No meaningful explanatory body
Why it was queued
cryptography-sensitive path
AI analysis · Informational 15/100
This commit is purely a build-system and test-harness cleanup. It refactors CMakeLists.txt, updates comments, simplifies the speculos test bridge, and removes a workaround in a test file. Nothing here changes the actual Ledger Bitcoin app that runs on the device, so it cannot directly affect user funds or device security.
Security candidateAdd unit tests for crypto_{get_key_fingerprint,master_key_fingerprint,derive_symmetric_key_slip21} and get_extended_pubkey_at_pathby Salvatore Ingala · 65279632 · May 29, 2026 · 1 fileMessage 60 · AdequateTriage 15Details
Commit message · Salvatore Ingala
Add unit tests for crypto_{get_key_fingerprint,master_key_fingerprint,derive_symmetric_key_slip21} and get_extended_pubkey_at_path
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
cryptography-sensitive path
Security candidateAdd initial mock for the dispatcher, and tests for get_preimageby Salvatore Ingala · 5ac1ec68 · May 19, 2026 · 8 filesMessage 60 · AdequateTriage 15Details
Commit message · Salvatore Ingala
Add initial mock for the dispatcher, and tests for get_preimage
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
cryptography-sensitive path
Security candidateNits from reviewby Salvatore Ingala · 9e23b56d · May 19, 2026 · 6 filesMessage 28 · OpaqueTriage 15Details
Commit message · Salvatore Ingala
Nits from review
28/100 · OpaqueMessage clarity
✓ Subject identifies a change! No meaningful explanatory body! Opaque security-relevant change
Why it was queued
cryptography-sensitive path
Security candidateAdditional unit testsby Salvatore Ingala · e130db4d · May 19, 2026 · 6 filesMessage 73 · AdequateTriage 15Details
Commit message · Salvatore Ingala
Additional unit tests
This reaches 100% coverage, except a few lines that can't be covered (either gcov artifacts, or non-reachable code because of defense-in-depth checks).
73/100 · AdequateMessage clarity
✓ Subject identifies a change✓ Provides detailed explanatory context✓ Explains rationale or failure mode✓ Mentions testing or verification