registry / @embarkai/ui-kit / 0.6.15
@embarkai/ui-kit@0.6.15
React UI components and hooks for embarkAI authentication and Account Abstraction
AI Security Review
scanned 3d ago · by lpm-firewall-aiThe inspected credential flows are tied to authentication and on-ramp operations and target embarkAI service endpoints. The manifest has no install lifecycle hooks, and iframe requests are dispatched through validated messages.
Static reason
High-risk behavior combination matched malicious policy.
Trigger
A consumer invokes authentication or an on-ramp query through the UI kit.
Impact
The reviewed behavior sends credentials to the package's own service for the requested feature. No unrelated collection or automatic install-time action was found.
Mechanism
The SDK passes its access token with specific iframe messages, which make requests to the configured embarkAI service endpoints; the iframe validates message origins.
AI rationale
The inspected credential flows are tied to authentication and on-ramp operations and target embarkAI service endpoints. The manifest has no install lifecycle hooks, and iframe requests are dispatched through validated messages.
Network endpoints2
api.embarkai.ioauth.embarkai.io
Decision evidence
public snapshotAI called this Clean at 96.0% confidence as Benign with low false-positive risk.
Evidence for AI clean decision
- The package describes itself as React UI components and hooks for embarkAI authentication and account abstraction.
- The manifest defines no install, preinstall, or postinstall lifecycle scripts.
- The service URLs point to embarkAI's own API and authentication hosts.
- On-ramp requests pass the access token to the embarkAI API for specific message types, and the SDK sends those messages when invoking the corresponding query.
- The iframe checks message origins before dispatching requests.
Evidence against
Behavioral surface
EnvironmentVarsEvalFilesystemNetwork
HighEntropyStringsUrlStrings
NoLicense
Source & flagged code
13 flaggeddist/iframe/assets/main-BBacRHFa.jsView file
68const [path, search = ""] = pathWithQuery.split("?", 2);
L69: const query = new URLSearchParams(search);
L70: const route = ACCOUNT_ROUTES.find((r) => r.method === method && r.path.test(path));
...
L245: function defaultTssUrl$2() {
L246: return "https://api.embarkai.io/tss";
L247: }
...
L382: if (response.status === 401) {
L383: if (await response.json().catch(() => null)) {
L384: this.clear();
...
L586: *
L587: * @private
L588: * @param {*} value The value to query.
Critical
Hardcoded Runtime Data Exfiltration
Source sends credentials or rich application records to a package-controlled external receiver enabled by default.
dist/iframe/assets/main-BBacRHFa.jsView on unpkg · L686323imports.wbg.__wbg_newnoargs_105ed471475aaf50 = function(arg0, arg1) {
L6324: return new Function(getStringFromWasm0(arg0, arg1));
L6325: };
High
Eval
Package source references dynamic code evaluation.
dist/iframe/assets/main-BBacRHFa.jsView on unpkg · L63234272patternName = generic_password
severity = medium
line = 4272
Medium
Findings
3 Critical3 High10 Medium5 Low
CriticalHardcoded Runtime Data Exfiltrationdist/iframe/assets/main-BBacRHFa.js
CriticalCredential Exfiltrationdist/index.js
CriticalTrigger Reachable Dangerous Capabilitydist/index.cjs
HighEvaldist/iframe/assets/main-BBacRHFa.js
HighTrigger Reachable Credential Exfiltrationdist/index.cjs
HighSemantic Analysis Limiteddist/styles.css
MediumSecret Patterndist/iframe/assets/main-BBacRHFa.js
MediumNetwork
MediumEnvironment Vars
MediumShips Wasm Moduledist/iframe/assets/dkls23_wasm_bg-Cn3nHBRm.wasm
MediumStructural Risk Force Deep Review
MediumSecret Patterndist/iframe/assets/main-BBacRHFa.js
Affected versions and remediation
This report applies to @embarkai/ui-kit@0.6.15.
See all versions of @embarkai/ui-kit for other recorded verdicts.
- Review the evidence and your use of @embarkai/ui-kit@0.6.15 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.