registry / @embarkai/ui-kit / 0.4.41
@embarkai/ui-kit@0.4.41
React UI components and hooks for EmbarkAI authentication and Account Abstraction
AI Security Review
scanned 4mo ago · by lpm-firewall-aiScanner findings appear to be false positives from wasm-bindgen glue, CSP allowances, local WASM loading, and legitimate product/API integrations. Static source inspection found no concrete malicious behavior.
Static reason
High-risk behavior combination matched malicious policy.; previous stored version diff introduced dangerous source
Trigger
User imports package or opens the packaged auth iframe during normal wallet/auth flows.
Impact
No unconsented execution, exfiltration, persistence, or destructive behavior identified.
Mechanism
Benign browser UI, auth, wallet, backup, and crypto helper code.
AI rationale
Scanner findings appear to be false positives from wasm-bindgen glue, CSP allowances, local WASM loading, and legitimate product/API integrations. Static source inspection found no concrete malicious behavior.
Evidence
package.jsondist/index.cjsdist/index.jsdist/iframe/main.jsdist/iframe/index.htmldist/iframe/_headersdist/iframe/oauth/x.jsdist/iframe/oauth/telegram.js
Decision evidence
public snapshotAI called this Clean at 88.0% confidence as Benign with low false-positive risk.
Evidence for AI clean decision
Evidence against
- package.json has no install/preinstall/postinstall lifecycle hooks; entries are dist/index.cjs and dist/index.js.
- dist/iframe/main.js WebAssembly load fetches local dkls23_wasm_bg.wasm via import.meta.url, not a remote decoded JS payload.
- dist/iframe/main.js new Function is wasm-bindgen glue (__wbg_newnoargs), not package-authored remote execution.
- Network calls are product-aligned: api.embarkai.io, dashboard.embarkai.io, auth.embarkai.io, Google Drive APIs, Telegram widget, Binance/Etherscan APIs.
- localStorage reads/writes are scoped to auth, passkey, keyshare backup, OAuth state, and wallet session functionality.
- No child_process import, lifecycle execution, credential harvesting loop, persistence mechanism, destructive action, or AI-agent control-surface writes found.
Behavioral surface
ChildProcessEnvironmentVarsEvalFilesystemNetwork
HighEntropyStrings
Source & flagged code
14 flaggeddist/iframe/main.jsView file
13__export(crypto_utils_exports, {
L14: base64ToBytes: () => base64ToBytes,
L15: bytesToBase64: () => bytesToBase64,
...
L156: return {
L157: data: bytesToBase64(new Uint8Array(encryptedData)),
L158: iv: bytesToBase64(iv),
...
L489: }
L490: const response = await fetch(`${TSS_URL}/api/auth/token/exchange`, {
L491: method: "POST",
...
L1910: }
L1911: async upload(fileName, content, usePrivateStorage = true) {
L1912: if (!this.isAuthenticated()) {
Critical
Remote Asset Decode Execute
Source fetches a remote non-code asset, decodes its contents, and dynamically executes the decoded payload.
dist/iframe/main.jsView on unpkg · L132710imports.wbg.__wbg_newnoargs_105ed471475aaf50 = function(arg0, arg1) {
L2711: const ret = new Function(getStringFromWasm0(arg0, arg1));
L2712: return ret;
High
334patternName = generic_password
severity = medium
line = 334
matchedText = throw ne...)));
Medium
Secret Pattern
Package contains a possible secret pattern.
dist/iframe/main.jsFindings
2 Critical2 High14 Medium5 Low
CriticalRemote Asset Decode Executedist/iframe/main.js
CriticalPrevious Version Dangerous Deltadist/index.cjs
HighChild Process
HighEvaldist/iframe/main.js
MediumSecret Patterndist/iframe/main.js
MediumNetwork
MediumEnvironment Vars
MediumShips Wasm Moduledist/iframe/dkls23_wasm_bg.wasm
MediumStructural Risk Force Deep Review
MediumSecret Patterndist/iframe/main.js
MediumSecret Patterndist/index.js
MediumSecret Patterndist/index.js
MediumSecret Patterndist/index.js
MediumSecret Pattern
Affected versions and remediation
This report applies to @embarkai/ui-kit@0.4.41.
See all versions of @embarkai/ui-kit for other recorded verdicts.
- Review the evidence and your use of @embarkai/ui-kit@0.4.41 before allowing it. Restrict the permissions described in this report.
- Choose an independently verified alternative or release. This report does not establish that other versions are safe.
Evidence last updated: .
Think this verdict is wrong? Dispute it on GitHub or read how reports are produced.