loadState() took hasWallet straight from storage, so any profile persisted
with the flag out of step with wallets stayed broken on every subsequent
load rather than only until the next write. The flag is now derived from
wallets.length at load time.
The only consumer is the popup's welcome-vs-wallet-list gate in
src/popup/index.js; the only writers set it true alongside an added
wallet. Nothing reads it expecting it to differ from wallets.length, so
deriving is safe and needs no write-back on load.
build.js records which emitted bundles contain src/shared/constants.js, and
constants.js carries a marker constant-folded from DEBUG itself. script/verify-build
cross-checks the two and fails on every way of not knowing, so deleting the
__BUILD_DEBUG__ define now breaks the build instead of shipping a live debug branch.
The password no longer crosses the extension messaging boundary: the popup
decrypts and signs, and sends only the raw signed transaction or the signature.
The background re-derives the signer from the artifact and checks it against
the approval it holds before broadcasting, so it is not a blind relay.
Fixes the highest-severity item in the repo: `src/shared/constants.js` had
`const DEBUG = true;`, so `generateMnemonic()` returned the publicly committed
`DEBUG_MNEMONIC` for every wallet created from a build of `main`, and the real
entropy path was dead code in every artifact we could produce.
## What changed
**`build.js`** — `AUTISTMASK_DEBUG` is read from the environment and injected
as a `__BUILD_DEBUG__` entry in the existing esbuild `define` map, next to the
other `__BUILD_*__` defines. Only the exact value `1` enables it; unset, empty,
`true`, or a typo all yield a release build, so the insecure direction requires
a deliberate opt-in and any mistake fails safe. The build prints
`Build mode: release (DEBUG off)` or
`Build mode: DEBUG (INSECURE - hardcoded test mnemonic, do not ship)`.
**`src/shared/constants.js`** — `DEBUG` now uses the same `typeof` guard that
`src/shared/buildInfo.js` already uses for the other build-time defines, and
defaults to `false` when the define is absent (jest, plain `require`).
`DEBUG_MNEMONIC` stays in the tree and stays exported.
**`Makefile`** — new `build-debug` target (`AUTISTMASK_DEBUG=1` + the same
build) so a debug build stays a one-liner for development.
**`README.md`** — new "Debug Builds" subsection under Getting Started, and the
DEBUG Mode Policy section now states that `DEBUG` is build-time-only and spells
out the boundary against the runtime toggle.
**`tests/wallet.test.js`** — new, covering both build modes.
**`TODO.md`** — refreshed in the same commit (details at the bottom).
No new `if (DEBUG)` branch was added and nothing about what DEBUG *does* changed:
still exactly the red banner plus the hardcoded test phrase, per the README
DEBUG Mode Policy and `RULES.md:76-80`.
## The interaction with the #145 settings toggle
This is the subtle part, so spelling out the reasoning.
There are two distinct debug flags in the tree after #145:
1. the compile-time `DEBUG` constant from `constants.js`, and
2. the runtime `debugMode` state flag, which the settings easter egg toggles
and which `settings.js:379` pushes into `log.js` via `setRuntimeDebug()`.
`log.js` merges them: `isDebug()` is `DEBUG || _runtimeDebug`. That merged
value feeds exactly two things — the log level threshold (`log.js:24`) and the
red banner (`views/helpers.js:71`). Making the banner user-toggleable is the
intended behavior of #145, and this PR leaves it alone.
`generateMnemonic()` does **not** consult `isDebug()`. It reads the
compile-time `DEBUG` binding directly. That distinction is what makes a release
build coherent: with `__BUILD_DEBUG__` false, `DEBUG` is false in the bundle,
so no amount of clicking the version ten times and flipping the toggle can
reach `return DEBUG_MNEMONIC`. The user can turn the banner and verbose logging
on in a release build; they cannot turn the hardcoded phrase on.
The failure mode to guard against is someone later "tidying up" the two flags by
routing `wallet.js` through `isDebug()`, which would silently reintroduce this
exact vulnerability with the runtime toggle as the trigger. Three things now
guard that: a comment at the `wallet.js` call site saying it must stay the
compile-time constant and why, the same statement in the README DEBUG Mode
Policy, and a regression test that calls `setRuntimeDebug(true)`, asserts
`isDebug()` is genuinely true, and then asserts `generateMnemonic()` still
returns fresh entropy.
I considered instead making the runtime toggle unavailable in release builds,
but rejected it: that removes a feature #145 deliberately added, and it defends
the wrong boundary. The banner is not the dangerous part; the mnemonic path is,
and that one is already unreachable.
## Verification
`make check` — green, 5 suites, 55 tests, plus lint and fmt-check. It also ran
via the pre-commit hook on the commit itself.
The new tests, per the verification standard in the manager comment on the
issue (not just `a !== b`) — with the flag off: two successive
`generateMnemonic()` calls differ, both pass `isValidMnemonic`, both are 12
words, neither equals `DEBUG_MNEMONIC`, and the result derives a usable HD
wallet (`xpub` + a well-formed first address), so a broken implementation
returning a counter or a truncated phrase would fail. Same assertions again
with the runtime toggle forced on. With the flag on (`__BUILD_DEBUG__` defined
before a `jest.resetModules()` re-require): `DEBUG` is `true` and
`generateMnemonic()` returns `DEBUG_MNEMONIC`, so the debug path is proven
working rather than silently deleted.
Build artifacts — `make build` and `make build-debug` both produce `dist/chrome`
and `dist/firefox` successfully. Grepping the minified bundles for the emitted
`DEBUG` export value across all four bundles (chrome popup, chrome background,
firefox popup, firefox background):
# after make build
$ grep -roh 'DEBUG:![01]' dist/chrome dist/firefox | sort | uniq -c
4 DEBUG:!1
# after make build-debug
$ grep -roh 'DEBUG:![01]' dist/chrome dist/firefox | sort | uniq -c
4 DEBUG:!0
`!1` is minified `false`, `!0` is `true`. Also checked the fail-safe path:
`AUTISTMASK_DEBUG=true make build` prints `Build mode: release (DEBUG off)` and
likewise yields `4 DEBUG:!1`.
One thing a reviewer should know about the grep: the `DEBUG_MNEMONIC` string
literal is still present in the release bundle. That is not a leak of anything
(the phrase is in this public repo already) and it does not mean the branch is
live — esbuild cannot tree-shake a CommonJS `module.exports` object, so the
constant survives while `DEBUG` folds to `false`. The compiled function is
`function PL(){return ML?UL:f_.fromEntropy(globalThis.crypto.getRandomValues(new Uint8Array(16))).phrase}`
where `ML` is the `DEBUG:!1` export. So "the phrase string is absent" is *not*
the right test for a release build; "the exported `DEBUG` is `!1`" is, which is
what I checked.
## `TODO.md` refresh
Per the manager comment: Status rewritten (no branch in flight —
`feat/issue-144-settings-about` landed as #145, scripts-to-rule-them-all landed
as #148, so the `scripts/` question is resolved; `make check` recorded as
verified green on `main` at `23aeae4`); the completed "Verify main passes make
check" Future Step removed; Future Steps rewritten against the #149-#168
backlog in rough priority order, keeping branch pruning (now #167) and the
pre-1.0 security review (noting #149 and #157 are parts of it but it is
broader).
One deliberate deviation to flag rather than bury: the manager asked that Next
Step become this issue. Taken literally against the Workflow section, this
commit *completes* #149, which would normally move it into Completed Steps. I
followed the repo's existing convention for in-flight work instead — the
previous Next Step was phrased as "Land feat/issue-144-settings-about", so Next
Step is now "Land #149 ... PR open, awaiting review", which is accurate until
this merges. Whoever merges should move it to Completed Steps and promote the
first Future Step. Happy to change it if the reviewer prefers the strict
reading.
## Out of scope
`script/lint` being `prettier --check` only and unable to catch undefined
identifiers (#152) — noted in the TODO but not fixed here; I greped for `DEBUG`
consumers by hand rather than relying on lint, as advised. Nothing else in the
DEBUG consumer set (`log.js`, `helpers.js`, `state.js`, `settings.js`) changed
behavior.
Co-authored-by: sneak <sneak@sneak.berlin>
Reviewed-on: #169
Co-authored-by: clawbot <clawbot@noreply.example.org>
Co-committed-by: clawbot <clawbot@noreply.example.org>
## Summary
Add a new well at the bottom of the settings view displaying application info (license, author, version, build date, git commit hash linked to Gitea), and an easter egg that reveals a debug mode toggle after clicking the version number 10 times.
## Changes
- **`src/popup/index.html`**: Added About well with license, author, version, build date, and linked commit hash. Added hidden debug well with debug mode toggle.
- **`src/popup/views/settings.js`**: Populate About well from build-time constants. Implement version click counter (10 clicks reveals debug well). Wire up debug mode toggle to state and runtime logger.
- **`src/shared/buildInfo.js`** (new): Module exporting build-time constants (`BUILD_VERSION`, `BUILD_LICENSE`, `BUILD_AUTHOR`, `BUILD_COMMIT`, `BUILD_DATE`, `GITEA_COMMIT_URL`) injected by esbuild define.
- **`build.js`**: Read git commit hash (short + full) and version from package.json, pass as esbuild `define` constants. Also reads `GIT_COMMIT_SHORT`/`GIT_COMMIT_FULL` env vars for Docker builds.
- **`Dockerfile`**: Accept `GIT_COMMIT_SHORT` and `GIT_COMMIT_FULL` build args, set as env vars for the build step.
- **`Makefile`**: Pass git commit hashes as Docker build args.
- **`src/shared/state.js`**: Add `debugMode` boolean to state (persisted).
- **`src/shared/log.js`**: Add runtime debug flag that supplements the compile-time `DEBUG` constant. Debug mode toggle updates this flag immediately.
closes#144
Co-authored-by: clawbot <clawbot@noreply.git.eeqj.de>
Co-authored-by: clawbot <clawbot@sneak.cloud>
Co-authored-by: user <user@Mac.lan guest wan>
Co-authored-by: clawbot <clawbot@eeqj.de>
Co-authored-by: sneak <sneak@sneak.berlin>
Reviewed-on: #145
Co-authored-by: clawbot <clawbot@noreply.example.org>
Co-committed-by: clawbot <clawbot@noreply.example.org>
## Summary
Fixes the view stack pop bug where pressing Back in Settings (or any view) always returned to Main instead of the previous view.
Closes [issue #134](#134)
## Problem
The popup UI had no navigation stack. Every back button was hardcoded to a specific destination (usually Main). The reported path:
> Main → Address → Transaction → Settings (gear icon) → Back
...would go to Main instead of returning to the Transaction view.
## Solution
Implemented a proper view navigation stack (like iOS) as already described in the README:
- **`viewStack`** array added to persisted state — survives popup close/reopen
- **`pushCurrentView()`** — pushes the current view name onto the stack before any forward navigation
- **`goBack()`** — pops the stack and shows the previous view; falls back to Main if the stack is empty; re-renders the wallet list when returning to Main
- **`clearViewStack()`** — resets the stack for root transitions (e.g., after adding/deleting a wallet)
### What Changed
1. **helpers.js** — Added navigation stack functions (`pushCurrentView`, `goBack`, `clearViewStack`, `setRenderMain`)
2. **state.js** — Added `viewStack` to persisted state
3. **index.js** — All `ctx.show*()` wrappers now push before navigating forward; gear button uses stack for toggle behavior
4. **All view back buttons** — Replaced hardcoded destinations with `goBack()` (settings, addressDetail, addressToken, transactionDetail, send, receive, addToken, confirmTx, addWallet, settingsAddToken, deleteWallet, export-privkey)
5. **Direct `showView()` forward navigations** — Added `pushCurrentView()` calls before `showView("send")` in addressDetail, addressToken, and home; before `showView("export-privkey")` in addressDetail; before `deleteWallet.show()` in settings
6. **Reset-to-root transitions** — `clearViewStack()` called after adding a wallet (all 3 import types), after deleting the last wallet, and after transaction completion (Done button)
### Navigation Paths Verified
- **Main → Settings → Back** → returns to Main ✓
- **Main → Address → Settings → Back** → returns to Address ✓
- **Main → Address → Transaction → Settings → Back** → returns to Transaction ✓ (the reported bug)
- **Main → Address → Token → Send → ConfirmTx → Back → Back → Back → Back** → unwinds correctly through each view back to Main ✓
- **Main → Address → Token → Transaction → Settings → Back** → returns to Transaction ✓
- **Settings → Add Wallet → (add) → Main** → stack cleared, fresh root ✓
- **Settings → Delete Wallet → Back** → returns to Settings ✓
- **Settings → Delete Wallet → (confirm)** → stack reset to [main], settings shown ✓
- **Address → Send → ConfirmTx → (broadcast) → SuccessTx → Done** → stack reset, returns to address context ✓
- **Popup close/reopen** → viewStack persisted, back navigation still works ✓
Co-authored-by: user <user@Mac.lan guest wan>
Reviewed-on: #146
Co-authored-by: clawbot <clawbot@noreply.example.org>
Co-committed-by: clawbot <clawbot@noreply.example.org>
## Summary
Fixes USD prices still showing on the main view when connected to a testnet (e.g. Sepolia). The root cause was stale mainnet prices lingering in the in-memory price cache after switching networks.
### Root Cause
PR #137 correctly made `refreshPrices()` skip fetching on testnets, but the cached prices from a prior mainnet session remained in the `prices` object. All display functions (`getPrice()`, `getAddressValueUsd()`, etc.) used whatever was cached without checking which network was active.
### Changes
- **`src/shared/prices.js`**
- `refreshPrices()` now clears the price cache when on a testnet instead of silently returning
- New `clearPrices()` function empties the cache and resets the fetch timestamp
- `getPrice()` returns null on testnets (defense-in-depth)
- `getAddressValueUsd()`, `getWalletValueUsd()`, `getTotalValueUsd()` return null on testnets
- **`src/popup/views/settings.js`**
- Network switcher immediately clears prices when switching to a testnet, so the UI updates without waiting for the next refresh cycle
closes#139
Co-authored-by: clawbot <clawbot@noreply.git.eeqj.de>
Reviewed-on: #142
Co-authored-by: clawbot <clawbot@noreply.example.org>
Co-committed-by: clawbot <clawbot@noreply.example.org>
## Summary
Adds Sepolia testnet support to AutistMask.
### Changes
- **New `src/shared/networks.js`** — centralized network definitions (mainnet + Sepolia) with chain IDs, default RPC/Blockscout endpoints, and block explorer URLs
- **State management** — `networkId` added to persisted state; defaults to mainnet for backward compatibility
- **Settings UI** — network selector dropdown lets users switch between Ethereum Mainnet and Sepolia Testnet
- **Dynamic explorer links** — all hardcoded `etherscan.io` URLs replaced with dynamic links from the current network config (`sepolia.etherscan.io` for Sepolia)
- **Background service** — `wallet_switchEthereumChain` now accepts both mainnet (0x1) and Sepolia (0xaa36a7); broadcasts `chainChanged` to connected dApps
- **Inpage provider** — fetches chain ID on init and updates dynamically via `chainChanged` events (no more hardcoded `0x1`)
- **Blockscout API** — uses `eth-sepolia.blockscout.com/api/v2` for Sepolia
- **Etherscan labels** — phishing/scam checks use the correct explorer per network
- **Price fetching** — skipped on testnets (testnet tokens have no real market value)
- **RPC validation** — checks against the selected network's chain ID, not hardcoded mainnet
- **ethers provider** — `getProvider()` uses the correct ethers `Network` for Sepolia
### API Endpoints Verified
| Service | Mainnet | Sepolia |
|---------|---------|--------|
| Etherscan | etherscan.io | sepolia.etherscan.io |
| Blockscout | eth.blockscout.com/api/v2 | eth-sepolia.blockscout.com/api/v2 |
| RPC | ethereum-rpc.publicnode.com | ethereum-sepolia-rpc.publicnode.com |
| CoinDesk (prices) | ✅ | N/A (skipped on testnet) |
closes#110
Reviewed-on: #137
THIS WAS ONESHOTTED USING OPUS 4. WTAF
Co-authored-by: clawbot <clawbot@noreply.example.org>
Co-committed-by: clawbot <clawbot@noreply.example.org>
Expands the confirm-tx warning system with three new warning types, all using the existing `visibility:hidden/visible` pattern from PR #98 (no animations, no layout shift).
## Changes
1. **Scam address list expanded** (7 → 652 addresses): Sourced from [MyEtherWallet/ethereum-lists](https://github.com/MyEtherWallet/ethereum-lists) darklist (MIT license). Checked synchronously before sending.
2. **Contract address warning**: When the recipient is a smart contract (detected via `getCode`), shows a warning that sending directly to a contract may result in permanent loss of funds.
3. **Null/burn address warning**: Detects known burn addresses (`0x0000...0000`, `0x...dead`, `0x...deadbeef`) and warns that funds are permanently destroyed.
4. **No-history warning** (existing from #98): Unchanged, still shows for EOAs with zero transaction history.
All warnings use reserved-space `visibility:hidden/visible` elements — no layout shift, no animations.
closes#114
Co-authored-by: clawbot <clawbot@noreply.git.eeqj.de>
Co-authored-by: user <user@Mac.lan guest wan>
Co-authored-by: clawbot <clawbot@eeqj.de>
Reviewed-on: #118
Co-authored-by: clawbot <sneak+clawbot@sneak.cloud>
Co-committed-by: clawbot <sneak+clawbot@sneak.cloud>
Fix multi-step Uniswap swap decoding and transaction display:
1. uniswap.js: In multi-step swaps (e.g. V3 → V4), the output token and
min received amount now come from the LAST swap step instead of the
first. Previously, an intermediate token's amountOutMin (18 decimals)
was formatted with the final token's decimals (6), producing
astronomically wrong 'Min. received' values (~2 trillion USDC).
2. transactions.js: Contract call token transfers (swaps) are now
consolidated into the original transaction entry instead of creating
separate entries per token transfer. This prevents intermediate hop
tokens (e.g. USDS in a USDT→USDS→USDC route) from appearing as the
transaction's Amount. The received token (swap output) is preferred.
3. transactions.js: The original transaction's from/to addresses are
preserved for contract calls, so the user sees their own address
instead of a router or Permit2 contract address.
closes#127
All isoDate() functions now output proper ISO 8601 format with timezone
offset (e.g. 2026-02-28T15:30:00-08:00) instead of bare datetime strings.
Also uses 'T' separator per ISO 8601.
closes#116
Add theme preference (light/dark/system) stored in extension state.
System mode follows prefers-color-scheme and listens for changes.
Dark mode inverts the monochrome palette (white-on-black).
Theme selector added to Display section in settings.
Closes#125
hdWalletFromXprv() and getSignerForAddress() for xprv type were deriving
addresses directly from the root key (m/N) instead of the standard BIP44
Ethereum path (m/44'/60'/0'/0/N). This caused imported xprv wallets to
generate completely wrong addresses.
Navigate to the BIP44 Ethereum derivation path before deriving child
addresses, matching the behavior of mnemonic-based wallet imports.
Add the ability to import an existing HD wallet using an extended
private key (xprv) instead of a mnemonic phrase.
- New 'xprv' wallet type with full HD derivation and address scanning
- New importXprv view with password encryption
- Updated getSignerForAddress to handle xprv wallet type
- Added xprv link to the add-wallet view
- Allow adding derived addresses for xprv wallets
Closes#20
- Add underline + click-to-copy (data-copy) to addresses in toAddressHtml()
so they match the style used everywhere else in the extension
- Fix 'USDT ETH' display: add rawValue to Uniswap decoder Amount details
and extract Token In info for proper symbol resolution in approval.js
- Hide redundant top-level Amount/To when decoded details are present
(they already show the same info inside the decoded section)
- Wrap decoded calldata details in a bordered well for visual separation
A single transaction (e.g. a DEX swap) can produce multiple ERC-20
token transfers. The transaction merger was keyed by tx hash alone,
so only the last token transfer survived. This meant the address-token
view's contract-address filter often matched nothing.
Use a composite key (hash + contract address) so all token transfers
are preserved. Also remove the bare normal-tx entry when it gets
replaced by token transfers to avoid duplicates.
Closes#72
When tokenBalances doesn't contain an entry for a token (e.g. before
balances are fetched), the symbol fell back to '?' in addressToken
and send views.
Add resolveSymbol() helper that checks tokenBalances → TOKEN_BY_ADDRESS
(known tokens) → trackedTokens → truncated address as last resort.
Fixes USDC and other known tokens showing '?' when balance data
hasn't loaded yet.
The Uniswap Universal Router calldata decoder listed V4_SWAP (0x10) in
command names but never decoded its inner actions to extract token
addresses. This caused all V4 swaps (e.g. USDT→USDC) to display as
'Swap ETH → ETH' because tokenIn/tokenOut defaulted to null, which
tokenInfo() resolved as native ETH.
Added decodeV4Swap() which parses the V4 inner action bytes:
- SETTLE (0x0b) → extracts input token currency address
- TAKE (0x0e) → extracts output token currency address
- SWAP_EXACT_IN/OUT and their SINGLE variants → extracts tokens from
pool keys and paths, plus amounts
These addresses are then resolved against the known token list
(TOKEN_BY_ADDRESS) to display correct symbols like USDT, USDC, etc.
Fixes#59
When a token's balance entry is missing or incomplete (e.g. not yet
fetched from Blockscout), the address-token view and send view now
fall back to the built-in known token list for symbol, name, and
decimals instead of showing '?'.
Also includes token name in the balance object returned by
fetchTokenBalances so the contract info well can display it.
Fixes#51
Two issues that could cause ENS names to disappear:
1. refreshBalances: on ENS lookup error, addr.ensName was set to null,
wiping any previously resolved name. Now keeps the existing value
on error — only overwrites on successful lookup.
2. ens.js cache: failed lookups were cached as null for 12 hours,
preventing retries even after transient errors resolved. Now skips
caching on failure so subsequent lookups retry immediately.
Added debug logging to ENS reverse lookups in refreshBalances.
- Map known DEX methods (execute, swap, multicall, etc.) to "Swap"
label instead of raw method name like "Execute"
- Remove unused displayData variable in transactionDetail.js
Addresses review feedback on PR #10.
- Preserve contract call metadata (direction, label, method) when token
transfers merge with normal txs in fetchRecentTransactions
- Handle 'contract' direction in counterparty display for home and
address detail list views
- Add decoded calldata display to transaction detail view, fetching
raw input from Blockscout and using decodeCalldata from approval.js
- Show 'Unknown contract call' with raw hex for unrecognized calldata
- Export decodeCalldata from approval.js for reuse
Names fetched from CoinGecko bulk API (100% coverage).
URLs sourced from ethereum-lists GitHub repo + manual curation
for major tokens. Some lesser-known tokens have empty URLs which
can be populated incrementally.
- Display full-precision amount (no 4-decimal truncation) in the
transaction detail view, with native quantity (wei/base units) below
- Both amount and native quantity are click-copyable
- Show wallet/address title above from/to when the address is ours
- Update README Display Consistency to document the exception
Replace stub error handlers with full approval flow for personal_sign,
eth_sign, eth_signTypedData_v4, and eth_signTypedData. Uses toolbar
popup only (no fallback window) and keeps sign approvals pending across
popup close/reopen cycles so the user can respond via the toolbar icon.
Token balances from Blockscout are now filtered before display.
A token only appears if it meets at least one criterion:
- In the known 511-token list (by contract address)
- Explicitly tracked by the user (added via + Token)
- Has >= 1,000 holders on-chain
Also rejects tokens spoofing a known symbol from a different
contract address (same check used for transaction filtering).
This prevents airdropped spam tokens like "OpenClaw" from
appearing in the wallet without the user ever tracking them.
Contract interactions (approve, swap, etc.) now display the method
name and token symbol instead of the meaningless 0 ETH value.
Blockscout provides the method name and whether the target is a
contract — parseTx uses these plus TOKEN_BY_ADDRESS to produce
labels like "Approve USDT" or "Swap LINK".
Added directionLabel field to parsed transactions so renderers
don't need to know about the sent/received/contract distinction.
Also: clicking a transaction on the home screen now opens the
transaction detail view instead of navigating to the address
detail view.
The dust filter was hiding contract interactions (approve, transfer,
etc.) because they have 0 ETH value, which falls below the dust
threshold. Contract calls with 0 ETH are normal — only plain ETH
transfers should be checked against the dust threshold.
Also captures is_contract and method from Blockscout's transaction
response for future use in transaction display.
Delete src/shared/tokens.js and migrate all consumers to
src/shared/tokenList.js which has 511 tokens (vs ~150) sourced
from CoinGecko with on-chain verified decimals.
- prices.js: getTopTokenPrices now from tokenList
- transactions.js: KNOWN_SYMBOLS now from tokenList (3.4x more
symbols for spoof detection)
- send.js: KNOWN_SYMBOLS for token dropdown filtering
- approval.js: uses pre-built TOKEN_BY_ADDRESS map instead of
constructing its own from TOKENS array
- addToken.js: uses getTopTokens(25) for quick-pick buttons
(only top 25 shown, not all 511)
New module with top Ethereum mainnet ERC-20 tokens sourced from
CoinGecko API (market cap ranking) with decimals verified on-chain
via eth_call to each contract. Addresses are EIP-55 checksummed.
Exports:
- TOKENS: full array ordered by market cap
- TOKEN_BY_ADDRESS: Map of lowercase address -> token info
- KNOWN_SYMBOLS: Map of uppercase symbol -> legitimate address
- getTopTokens(n): return first n tokens
This module is not yet integrated into the existing token.js — it
exists alongside it for now as a data source ready for integration.
When sending an ERC-20 token, Blockscout returns the same tx hash
from both the /transactions endpoint (as a 0 ETH contract call) and
the /token-transfers endpoint (as the actual token transfer with
amount and symbol). The old dedup logic kept the 0 ETH version and
skipped the token transfer.
Now token transfers replace normal transactions with the same hash,
since the token transfer has the real amount, symbol, and contract
address.
Creates a centralized transactionDetail.js view module, replacing
the duplicated showTxDetail/copyableHtml/blockieHtml/txDetailAddressHtml
code that was in both addressDetail.js and addressToken.js (~120 lines
removed). Transaction data is stored in state.viewData and persisted,
so the transaction detail view survives popup close/reopen.
Adds viewData to persisted state. Each view that needs data for
restore stores it in state.viewData before rendering. The ctx object
now has showTransactionDetail() alongside all other show methods.
Restorable views expanded to include: transaction (via viewData.tx),
success-tx (via viewData.hash/blockNumber), error-tx (via
viewData.message). txStatus.js split into show (sets data) + render
(reads data) for each screen, enabling restore.
Non-restorable views (send, confirm-tx, wait-tx, add-wallet,
import-key, add-token) fall back to the nearest parent since they
involve active form state or network polling.
The current view, selected wallet, selected address, and selected
token are now saved to extension storage. When the popup reopens,
it restores to the last visited view instead of always returning
to the home screen.
Restorable views: main, address detail, address-token, receive,
settings. Non-restorable views (send, confirm, tx status, forms)
fall back to the nearest parent. Stored indices are validated
against current wallet data to handle stale references.
Also refactors receive view setup into a centralized receive.show()
function, eliminating duplicate QR/address/warning code from
addressDetail.js, addressToken.js, and home.js. Adds settings.show()
to centralize settings field population.
Clicking a token balance on the address detail view navigates to a
focused view showing only that token's transactions. Send pre-selects
and locks the token dropdown, Receive shows an ERC-20 warning for
non-ETH tokens, and all back buttons return to the correct parent view.
Add debugFetch wrapper in log.js that logs method, URL, and body on
request, and status code on response. Replace all fetch() calls
across balances, transactions, tokens, background RPC proxy, and
settings validation with debugFetch.
Add showZeroBalanceTokens setting (default: on). When enabled,
balanceLinesForAddress merges state.trackedTokens with the address's
tokenBalances, showing 0.0000 lines for tracked tokens that have no
balance on that address. This gives users visibility into all tokens
they're watching across all addresses.
Address poisoning attacks also use real native ETH dust transfers
(e.g. 1 gwei) from look-alike addresses. Token-level filters cannot
catch these. Add a configurable dust threshold (default 100,000 gwei
/ 0.0001 ETH) that hides transactions below the threshold from
history. The threshold is editable in Settings and the filter can be
disabled entirely. Document the specific attack tx in the README.
Three layers of defense against address poisoning attacks:
1. Known symbol verification: tokens claiming a symbol from the
hardcoded top-250 list (e.g. "ETH", "USDT") but from an
unrecognized contract are identified as spoofs and always hidden.
Their contract addresses are auto-added to the fraud blocklist.
2. Low-holder filtering: tokens with <1000 holders are hidden from
both transaction history and the send token selector. Controlled
by the "Hide tokens with fewer than 1,000 holders" setting.
3. Fraud contract blocklist: a persistent local list of detected
fraud contract addresses. Transactions involving these contracts
are hidden. Controlled by the "Hide transactions from detected
fraud contracts" setting.
Both settings default to on and can be disabled in Settings.
Fetching and filtering are separated: fetchRecentTransactions returns
raw data, filterTransactions is a pure function applying heuristics.
Token holder counts are now passed through from the Blockscout API.
allowedSites and deniedSites are now objects keyed by address instead
of flat arrays, so approving a site for one address no longer grants
access for all addresses. Old flat-array data is discarded on load.
Settings view collects unique hostnames across all addresses and
deleting removes the site from every address.
Scanning: check all gap-limit addresses in parallel per batch instead
of sequentially. For a wallet with 1 used address this reduces from
12 sequential RPC round-trips to 1 parallel batch + 1 small follow-up.
Display: add shared formatAddressHtml(address, ensName, maxLen) and
escapeHtml() to helpers.js. Use them in confirm-tx (was missing color
dot entirely) and approval view. Remove duplicate escapeHtml from
addressDetail.js.
- Add activeAddress, allowedSites, deniedSites, rememberSiteChoice to
persisted state
- Replace auto-connect with permission checks: allowed sites connect
automatically, denied sites are rejected, unknown sites trigger an
approval popup
- Add approval popup UI with hostname display, active address preview,
remember checkbox, and allow/deny buttons
- Add ACTIVE/[select] indicator on address rows in the main view to
set the active web3 address
- Add allowed/denied site list management in settings with delete buttons
- Broadcast accountsChanged to connected dapps when active address changes
- Handle approval window close as implicit denial
Deterministic colored dots derived from address bytes (16-color palette)
displayed before every address. ENS reverse resolution for transaction
counterparties with 12-hour localStorage cache.
- Transaction values now use exactly 4 decimal places (was 6),
matching balance display everywhere else
- Transaction detail view shows "2026-02-25 15:04:23 (23 days ago)"
instead of just the ISO date
- Added Display Consistency policy to README
Blockscout v2 API rejects the `limit` query parameter on
/transactions and /token-transfers endpoints (returns 422).
Remove it and slice results client-side instead.
Major changes:
- Fetch token balances and tx history from Blockscout API (configurable)
- Remove manual token discovery (discoverTokens) in favor of Blockscout
- HD address gap scanning on mnemonic import
- Duplicate mnemonic detection on wallet add
- EIP-6963 multi-wallet discovery + selectedAddress updates in inpage
- Two-tier balance refresh: 10s while popup open, 60s background
- Fix $0.00 flash before prices load (return null when no prices)
- No-layout-shift: min-height on total value element
- Aligned balance columns (42ch address width, consistent USD column)
- All errors use flash messages instead of off-screen error divs
- Settings gear in global title bar, add-wallet moved to settings pane
- Settings wells with light grey background, configurable Blockscout URL
- Consistent "< Back" buttons top-left on all views
- Address titles (Address 1.1, 1.2, etc.) on main and detail views
- Send view shows current balance of selected asset
- Clickable affordance policy added to README
- Shortened mnemonic backup warning
- Fix broken background script constant imports
Split popup/index.js (784 lines) into focused modules:
- shared/state.js: state management, storage persistence
- shared/wallet.js: mnemonic gen, HD derivation, signing
- shared/prices.js: price cache (5min TTL), USD formatting,
value aggregation (address → wallet → total)
- shared/balances.js: ETH + ERC-20 balance cache (60s TTL),
ENS lookup, token contract metadata lookup
- shared/vault.js: unchanged (libsodium encryption)
- shared/tokens.js: unchanged (token list + CoinDesk client)
- popup/index.js: view switching and event wiring only
Token tracking is now app-wide: trackedTokens stored in state,
balances fetched for all tracked tokens across all addresses.
Add Token now calls the real contract to read name/symbol/decimals.
Total portfolio value shown in 2x type on Home screen.
