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 is a small bug-fix patch for Electrum's QML (mobile-style) graphical interface. The crash reporter was not correctly handling exceptions that come from Python's threading system, which could cause the crash reporter itself to misbehav…
Fixes incorrect exception argument handling in crash reporterPrevents potential secondary failure in error-reporting pathNo security boundary crossed; no input from untrusted source is processed
This commit fixes a bug in Electrum's mobile-style QML crash reporter. When an uncaught exception occurred inside a background thread, the crash reporter itself would break because it expected three pieces of information but received one b…
Crash reporter failure could hide or obscure real runtime errors in background threadsNo input validation, privilege boundary, or cryptographic issue presentFix improves diagnostic fidelity but does not close an attack vector
This commit changes how Electrum creates unsigned Bitcoin transactions when an existing 'base transaction' is provided. Previously, the wallet wrongly demanded spendable coins even when the caller supplied a base transaction that already c…
Relaxation of input-source preconditionAddition of consensus-level invariant assertion (at least one input)Addition of empty-input guard before PartialTransaction construction
This commit fixes a crash in Electrum's transaction builder. If a user had only one on-chain coin (UTXO) that was just enough to meet a reserve requirement, clicking 'Max' in the send tab would hit an internal assertion and fail instead of…
Denial-of-service-like crash path triggered by user action in normal wallet stateAssertion failure reachable through GUI send flowFix converts assertion crash into handled NotEnoughFunds exception
This is a routine maintenance commit that updates many of Electrum's software building blocks (dependencies) and raises the minimum required Python version from 3.10 to 3.11. It also refreshes the Linux distribution snapshots used to build…
Dependency version bumps for security-sensitive libraries (cryptography, cffi, libffi, sqlite)Removal of an embedded PGP public key and switch to hash verification for Windows Python installer downloadsBase image refresh from EOL/near-EOL Debian bullseye to bookworm for AppImage builds
This commit is a straightforward internal cleanup: it moves repeated checks like 'are we running on Windows?', 'are we on Android?', and 'are we inside an AppImage?' into a single new helper module called platform.py. The actual behavior o…
This commit is a straightforward internal cleanup: it moves repeated operating-system checks (Windows, macOS, Linux, Android, AppImage) into a single new helper module called platform.py and updates the rest of the code to use those helper…
This commit only changes the software version number from 4.8.3 to 4.8.4 in a single file. It contains no security-related changes, no bug fixes, and no functional code modifications.
This commit relaxes a wallet check when creating a new Bitcoin transaction. Previously, Electrum refused to create a transaction if the user had no spendable coins and no explicit inputs. The change allows creation to continue if a 'base t…
Relaxation of input/coin availability guardAddition of post-construction input count assertionPotential workflow: Replace-By-Fee (RBF) or transaction bumping using base_tx inputs
This commit is a code cleanup in Electrum's mobile-style QML wallet wizard. It replaces an older thread-based background scan for BIP39 seed accounts with a newer asyncio-based task approach, and makes the recovery dialog destroy itself wh…
Refactor of async task lifecycle and cancellation handlingDialog lifecycle cleanup with onClosed: destroy()Use of ignore_if_destroyed to avoid callbacks on destroyed QML objects
This commit is a code cleanup in Electrum's mobile-style QML wallet wizard. It replaces an extra background thread with a simpler asyncio-based callback for scanning BIP39 seed accounts. There is no indication this fixes a security bug; it…
This commit fixes a bug in Electrum's Qt 'Send' tab where selecting a contact from the autocomplete dropdown could briefly leave the wrong recipient active. Previously, the payment field was set to a display name like 'Alice <alice@example…
UI state desynchronization between displayed recipient and active PaymentIdentifierStale PaymentIdentifier usable for ~1 second after recipient text changesPay/Save/Max buttons enabled while recipient is being edited
This commit fixes a UI bug in Electrum's QML (mobile-style) interface where transaction and address labels could become stale or fail to update after editing. It replaces a fragile manual callback system with a proper event listener so the…
No security-relevant signals in diff: no privilege escalation, injection, cryptographic, or authentication changes.Fixes a UI consistency bug (stale labels) that could mislead users but does not enable direct exploitation.Adds event-listener lifecycle management (register/unregister on destroy) to prevent dangling callbacks and potential use-after-free patterns.
This commit fixes a bug in Electrum's mobile-style QML password dialog. If a user tapped the OK/confirm button while the dialog was already closing, the app could emit an extra 'password entered' signal. That extra signal triggered a crash…
Crash in authentication-related code path (AuthMixin)Duplicate signal emission during UI state transitionUser-interaction race condition (tap while dialog closing)
This commit is a performance fix for a wallet function that traces where a transaction's funds came from. It replaces a slow, repeatedly-recursive approach with a more efficient loop and removes an old cache. The change also fixes how the …
Performance fix for wallet history traversalAddress-reuse uncle detection logic changedCache removed in favor of iterative computation
This commit fixes a bug where Electrum wallets created with unusual, hand-crafted 'old-style' seed phrases could not be opened or used. These seeds decode to an odd number of hex characters, which the program previously rejected. The chang…
Input validation relaxed for legacy data (odd-length hex strings)New user-facing error path prevents seed display when mnemonic encoding is impossibleGUI exception handling narrowed: UserFacingException shown as str, other Exception as repr
This commit is a small follow-up cleanup. It translates several user-facing error messages into different languages and makes the Qt GUI handle one specific file-read/write error (StorageReadWriteError) more gracefully, showing a warning d…
Exception-handling path changed in GUIUser-facing error messages internationalizedNo changes to encryption, wallet parsing semantics, or permissions
This commit improves Electrum's error handling when a user tries to open a file that is not a valid Electrum wallet. Previously, if the file contained bytes that were not valid UTF-8 text, the program would crash with a confusing technical…
Error-handling hardeningUser-facing exception message improvedNo cryptographic or parsing logic changed
This is a small user-interface fix in Electrum's Qt desktop app. Previously, if the wallet file was on a read-only folder or an unplugged drive, the app treated the resulting StorageReadWriteError as an unexpected crash and sent a crash re…
No memory safety, cryptographic, or authorization changesException handling change only; no new attack surface introducedCrash-reporting no longer triggered for expected I/O failure scenarios
This commit is a small follow-up cleanup in Electrum's keystore code. It renames an internal helper function, makes the code accept certain unusual 'old' seed phrases that some users created by hand, improves an error message shown to user…
Refactoring of seed validation helperUser-facing error handling improvement in seed displayGrandfathering of non-standard old seed phrases
Merge pull request #10988 from f321x/update_security_review_ci_model
ci: bump security review model to opus 5.5, detect downgrades
78/100 · AdequateMessage clarity
✓ Specific, descriptive subject✓ Names a concrete action or component✓ Provides an explanatory body✓ Links an issue, advisory, or supporting reference✓ Names security-relevant behavior explicitly
Why it was queued
update trustmerge-commit duplicate discount
AI analysis · Informational 15/100
This commit updates Electrum's internal CI (continuous integration) script that runs an automated security review using Anthropic's Claude Code tool. It changes the AI model version used for reviews from 'claude-opus-5' to 'claude-opus-5-5', switches the output format to a streaming JSON format, and adds detection and warning logic for cases where Claude Code silently falls back to a different (presumably weaker) model. There is no change to Electrum wallet code, cryptography, networking, or user-facing behavior.
Lower-priorityci: bump security review ci model to opus 5.5by f321x · f4d0745d · Sep 23, 2026 · 1 fileMessage 77 · AdequateTriage 0Details
Commit message · f321x
ci: bump security review ci model to opus 5.5
Bump the LLM used by the security review CI script from Claude Opus 5 to Claude Opus 5.5
77/100 · AdequateMessage clarity
✓ Descriptive subject✓ Names a concrete action or component✓ Uses a recognizable type or scope✓ Provides an explanatory body✓ Names security-relevant behavior explicitly
Security candidateci: security review: warn if the model got downgradedby f321x · ace2ca7a · Sep 23, 2026 · 1 fileMessage 90 · StrongInformational 12Details
Commit message · f321x
ci: security review: warn if the model got downgraded
Use stream-json output to detect when Claude Code falls back to another model (e.g. after a cyber safeguards refusal), and show a warning in the CI log, as a GitHub Actions annotation, and in the PR comment.
90/100 · StrongMessage clarity
✓ Specific, descriptive subject✓ Names a concrete action or component✓ Uses a recognizable type or scope✓ Provides detailed explanatory context✓ Names security-relevant behavior explicitly
Why it was queued
update trust
AI analysis · Informational 12/100
This commit updates Electrum's own CI security-review script. It adds detection and warnings when the automated code reviewer (Claude Code) silently falls back to a different AI model, for example after a content-policy refusal. It does not change Electrum wallet code, network behavior, cryptography, or anything end users install. It is purely an internal CI/quality-assurance improvement.
transaction: treat native segwit input with non-empty witness as complete
68/100 · AdequateMessage clarity
✓ Descriptive subject✓ Names a concrete action or component✓ Provides an explanatory body✓ Links an issue, advisory, or supporting reference
Why it was queued
signing or wallet pathmerge-commit duplicate discountsecond-pass: security-sensitive path
AI analysis · Low 44/100
This commit fixes a bug where Electrum failed to recognize certain already-signed Bitcoin transactions as complete. Specifically, for native SegWit inputs, Bitcoin Core can produce a finalized PSBT with a valid witness but an empty scriptSig. Electrum's 'is this transaction complete?' check previously required a non-empty scriptSig for such inputs, so it wrongly treated these valid, signed transactions as incomplete. The fix adds a rule: if a native SegWit input has a non-empty witness, treat it as complete. This could have led users to try re-signing or re-broadcasting, or to confusion about whether a transaction was ready.
AI review queuedtransaction: treat native segwit input with non-empty witness as completeby ThomasV · e9bb42e6 · Sep 22, 2026 · 2 filesMessage 81 · StrongLow 35Details
Commit message · ThomasV
transaction: treat native segwit input with non-empty witness as complete
Bitcoin Core, Sparrow omit the final scriptSig is empty, for native segwit inputs, and only write the witness. PartialTxInput.is_complete() should not require both fields to be present. fixes #10986
81/100 · StrongMessage clarity
✓ Specific, descriptive subject✓ Names a concrete action or component✓ Provides detailed explanatory context✓ Links an issue, advisory, or supporting reference
Why it was queued
signing or wallet pathsecond-pass: security-sensitive path
AI analysis · Low 35/100
This commit fixes a bug where Electrum incorrectly treated finalized Bitcoin transactions as incomplete. Some wallet software (like Bitcoin Core and Sparrow) creates native SegWit transactions that omit an empty placeholder field Electrum expected. As a result, Electrum could wrongly refuse to broadcast or mishandle valid, already-signed transactions. The fix makes Electrum recognize these transactions as complete when the witness data is present.
Merge pull request #10985 from SomberNight/202609_fxrate_mempoolspace
exchange_rate: add mempool.space
58/100 · ThinMessage clarity
✓ Specific, descriptive subject✓ Names a concrete action or component✓ Links an issue, advisory, or supporting reference! No meaningful explanatory body
- it has both spot and historical prices for 7 major ccys - though historical prices before 2022-05 only have weekly resolution, which does not work with the rest of our code (which expects daily data)
lightning: merge LocalConfig and RemoteConfig into ChannelConfig
validate_params takes a is_local parameter for the stricter checks we run on our own config.
65/100 · AdequateMessage clarity
✓ Specific, descriptive subject✓ Names a concrete action or component✓ Provides an explanatory body
Why it was queued
defensive validation
Security candidatelightning: give LocalConfig and RemoteConfig the same fields.by ThomasV · 41ba9ce5 · Sep 21, 2026 · 7 filesMessage 73 · AdequateLow 27Details
Commit message · ThomasV
lightning: give LocalConfig and RemoteConfig the same fields.
Based on ab0cab8a1 in the 'peerbackups' branch. Both parties now have current_commitment_signature, current_htlc_signatures current and next per-commitment points. For REMOTE, our signatures are saved in sign_next_commitment.
Note that this creates a few extra disk writes per channel commitment. The missing fiedls are added in the same WalletDB upgrade as previous commit.
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 27/100
This commit refactors how Electrum stores Lightning channel configuration data so that both sides of a channel (local and remote) use the same set of fields. It adds per-commitment points and signature fields to the shared base class, updates channel setup and state-transition code to populate them, and includes a database upgrade for existing wallets. The change appears to be a structural cleanup and preparation for future peer-backup functionality rather than a direct fix for an active security bug. There is no explicit vendor statement that this is a security patch, and no independent researcher is credited.
Lower-priorityMerge pull request #10983 from f321x/fix_broken_type_hintby Felix · 0f631852 · Sep 21, 2026 · 1 fileMessage 58 · ThinTriage 0Details
Commit message · Felix
Merge pull request #10983 from f321x/fix_broken_type_hint
lnpeer: fix broken type hint
58/100 · ThinMessage clarity
✓ Specific, descriptive subject✓ Names a concrete action or component✓ Links an issue, advisory, or supporting reference! No meaningful explanatory body
Merge pull request #10981 from spesmilo/do_not_sign_txin_with_witness
transaction: do not sign txin if if already has a witness
73/100 · AdequateMessage clarity
✓ Specific, descriptive subject✓ Names a concrete action or component✓ Provides an explanatory body✓ Links an issue, advisory, or supporting reference
Why it was queued
signing or wallet pathmerge-commit duplicate discountsecond-pass: security-sensitive path
AI analysis · Moderate 68/100
This change prevents Electrum from re-signing Bitcoin transaction inputs that already contain a witness (the data proving the input was authorized). The patch fixes a bug where Electrum could incorrectly sign a non-SegWit input as if it were SegWit when a PSBT already included a witness for that input. This could lead to producing invalid or unintended signatures during transaction signing, especially when handling partially-signed transactions from external signers.
AI review queuedlightning: derive local channel keys instead of storing them (see #4909)by ThomasV · 461238aa · Sep 19, 2026 · 9 filesMessage 81 · StrongLow 31Details
Commit message · ThomasV
lightning: derive local channel keys instead of storing them (see #4909)
Keep only pubkeys in LocalConfig. Privkeys and per_commitment_secret_seed are removed; they are re-derived from channel_seed on startup, and checked against the pubkeys. Secrets (channel_seed, multisig privkey) and funding_locked_received are moved to channel storage.
WalletDB conversion 73->74.
81/100 · StrongMessage clarity
✓ Specific, descriptive subject✓ Names a concrete action or component✓ Provides detailed explanatory context✓ Links an issue, advisory, or supporting reference
Why it was queued
signing or wallet pathsecond-pass: security-sensitive path
AI analysis · Low 31/100
This commit refactors how Electrum stores Lightning channel keys. Instead of keeping private keys inside the channel configuration object that gets saved to disk, it now keeps only public keys there and re-derives the private keys from a channel seed when the wallet starts up. The goal is to reduce the amount of secret material sitting in persisted channel data. It also adds a wallet database upgrade that strips old private keys from existing wallets and moves the remaining secrets into the channel storage. This is a defensive hardening change, not a fix for an active exploit.
Merge pull request #10974 from SomberNight/202609_nostr_psbt_xpub
plugins: psbt_nostr: document design choice for using xpub as nsec
73/100 · AdequateMessage clarity
✓ Specific, descriptive subject✓ Names a concrete action or component✓ Provides an explanatory body✓ Links an issue, advisory, or supporting reference
Why it was queued
signing boundarysigning or wallet pathmerge-commit duplicate discount
AI analysis · Informational 15/100
This commit only adds a long code comment explaining a design choice in Electrum's PSBT-over-Nostr plugin. It does not change any program behavior. The comment documents that the plugin derives each cosigner's Nostr secret key from their Bitcoin extended public key (xpub), which means every cosigner can also compute every other cosigner's secret key. The developers explicitly call this an accepted tradeoff because the only shared cryptographic material among cosigners is the xpubs, and the Nostr keys are only used to exchange proposed transaction drafts (PSBTs). No vulnerability is introduced or fixed by this change.
Lower-priorityplugins: manifest.json: "description" must now be plaintextby SomberNight · e2584017 · Sep 18, 2026 · 3 filesMessage 86 · StrongTriage 0Details
Commit message · SomberNight
plugins: manifest.json: "description" must now be plaintext
- rich text can contain embedded base64-encoded images - qt docs say [0] it is not safe to parse untrusted input as an image
> Security Considerations: > Since these file formats are more rarely used, the codecs may be > less thoroughly debugged against potential security holes. As always, > care should be taken when creating applications that may be used > to decode uncontrolled data files.
86/100 · StrongMessage clarity
✓ Specific, descriptive subject✓ Names a concrete action or component✓ Provides detailed explanatory context✓ Links an issue, advisory, or supporting reference✓ Names security-relevant behavior explicitly
Merge pull request #10980 from spesmilo/add_input_from_network_do_not_ignore_exceptions
add_info_from_network: do not swallow exceptions
73/100 · AdequateMessage clarity
✓ Specific, descriptive subject✓ Names a concrete action or component✓ Provides an explanatory body✓ Links an issue, advisory, or supporting reference
Why it was queued
signing or wallet pathmerge-commit duplicate discountsecond-pass: security-sensitive path
AI analysis · Low 42/100
This is a one-line bugfix in Electrum's transaction handling. Previously, the code accidentally swallowed (hid) almost all exceptions when fetching extra transaction data from the network, only re-raising errors that were both network-related and not set to be ignored. After the fix, any unexpected error is re-raised so it can be handled or reported properly. The change improves reliability and error visibility, but the diff alone does not show an active security vulnerability being exploited.
Merge pull request #10970 from f321x/fix_bolt11_bugs_2
bolt11: bugfixes and cleanup
58/100 · ThinMessage clarity
✓ Specific, descriptive subject✓ Names a concrete action or component✓ Links an issue, advisory, or supporting reference! No meaningful explanatory body
Why it was queued
signing or wallet pathmerge-commit duplicate discountsecond-pass: security-sensitive path
AI analysis · Moderate 60/100
This commit fixes several bugs in Electrum's handling of BOLT11 Lightning invoices. The most user-visible fixes are: stricter validation of invoice amounts (rejecting zero, negative, sub-millisatoshi, and leading-zero amounts), correct padding of the timestamp field so small dates don't get corrupted, preserving the 'n' (public key) tag when decoding so re-encoding doesn't silently drop it, and removing support for the non-standard 't' routing tag that could previously hide malformed data. There is also a fix to how description fields are truncated so they don't produce invalid UTF-8. These are correctness and robustness fixes rather than a single critical vulnerability, but they could have allowed malformed or ambiguous invoices to be accepted or re-encoded incorrectly.
Security candidateadd_info_from_network: do not swallow exceptionsby ThomasV · 3f67b21b · Sep 18, 2026 · 1 fileMessage 80 · StrongModerate 61Details
Commit message · ThomasV
add_info_from_network: do not swallow exceptions
An attacker can provide a witness_utxo whose value disagrees with the real previous transaction. That would raise PSBTInputConsistencyFailure, which is swallowed because it is not a NetworkException.
This would lead us to sign a tx with the wrong input amount. (no fund loss: the tx is invalid under BIP143)
80/100 · StrongMessage clarity
✓ Descriptive subject✓ Names a concrete action or component✓ Provides detailed explanatory context✓ Explains rationale or failure mode
Why it was queued
explicit security languagesigning boundarysigning or wallet path
AI analysis · Moderate 61/100
Electrum's transaction builder had a bug where it would ignore certain consistency errors when fetching extra transaction data from the network. A malicious or misbehaving server could supply a wrong input amount, and Electrum would silently continue and sign the transaction with that wrong amount. The resulting transaction would be invalid and not accepted by the Bitcoin network, so no funds could be lost, but it could cause confusion, failed payments, or be used to probe wallet behavior.
AI review queuedbolt11: remove tag arg from get_routing_infoby f321x · e78fba28 · Sep 18, 2026 · 7 filesMessage 60 · AdequateInformational 15Details
Commit message · f321x
bolt11: remove tag arg from get_routing_info
Through the removal of t-type routing info tags this argument isn't neccessary anymore.
60/100 · AdequateMessage clarity
✓ Descriptive subject✓ Names a concrete action or component✓ Provides an explanatory body
Why it was queued
signing or wallet pathsecond-pass: security-sensitive path
AI analysis · Informational 15/100
This is a small code cleanup change. A function that reads routing hints from Lightning invoices no longer takes a tag argument because only one type of routing hint ('r') is now supported. All callers are updated accordingly. There is no security issue visible in the change.
Merge pull request #10976 from SomberNight/202609_security_md_name
SECURITY.md: ask people to say how they want to be credited as
78/100 · AdequateMessage clarity
✓ Specific, descriptive subject✓ Names a concrete action or component✓ Provides an explanatory body✓ Links an issue, advisory, or supporting reference✓ Names security-relevant behavior explicitly
tooltips use QLabel internally, so due to InjectNoRichTextEventFilter, their textFormat is no longer AutoText, but is now PlainText instead.
note: messages.to_rtf() was specifically introduced for usage with tooltips [0], I guess to nudge Qt to break longer lines more often (so that tooltips widths are smaller). Tooltips being forced to PlainText breaks that. I also cannot see an easy way to opt-in the tooltip of e.g. a QCheckBox to RichText.
Instead, now we use the "textwrap" module from the stdlib to manually word-wrap the strings.
✓ Descriptive subject✓ Names a concrete action or component✓ Provides detailed explanatory context✓ Explains rationale or failure mode✓ Links an issue, advisory, or supporting reference