adbf@1.0.152
adbf
AI Security Review
scanned 23d ago · by lpm-firewall-aiThe source exposes a user-triggered arbitrary-destination request and header-forwarding mechanism inside an unusually opaque bundle. There is no install-time trigger or confirmed exfiltration sink, so blocking is not supported.
Static reason
High-risk behavior combination matched malicious policy.; previous stored version diff introduced dangerous source
Trigger
A user runs the adbf command and loads or submits the served page.
Impact
This is a potentially dangerous arbitrary-destination request capability, but static evidence does not establish automatic execution, credential theft, or a package-controlled recipient.
Mechanism
Browser-side code derives the submitted URL's origin, posts to its auth path, then fetches the submitted URL with a generated accept-user-agent header.
AI rationale
The source exposes a user-triggered arbitrary-destination request and header-forwarding mechanism inside an unusually opaque bundle. There is no install-time trigger or confirmed exfiltration sink, so blocking is not supported.
Evidence
index.js
Decision evidence
public snapshotAI called this Suspicious at 84.0% confidence as Dangerous Capability with medium false-positive risk.
Evidence for warning
- The manifest exposes index.js through the adbf command and contains no install lifecycle hook.
- The bundled page accepts a user-provided URL, derives its origin, requests an auth path on that origin, and sends the result as a custom header when fetching that URL.
- The entrypoint is heavily obfuscated, preventing a reliable benign-purpose assessment.
Evidence against
- No package lifecycle script automatically runs the entrypoint during installation.
- The observed network flow is activated through the command and page interaction, not automatically on import.
Behavioral surface
CryptoDynamicRequireEnvironmentVarsEvalFilesystem
HighEntropyStringsPossibleObfuscationUrlStrings
Source & flagged code
12 flagged · loading sourceindex.jsView file
1var a0_0x5a7c88=a0_0x443f;(function(_0x4aa430,_0x26977f){var _0x40c52e=a0_0x443f,_0xbf433b=_0x4aa430();while(!![]){try{var _0x2a9c0f=parseInt(_0x40c52e(0x2192))/0x1*(-parseInt(_0x4...
L2:
Critical
Command Output Exfiltration
Source executes local commands and sends command output to an external endpoint.
index.jsView on unpkg · L11var a0_0x5a7c88=a0_0x443f;(function(_0x4aa430,_0x26977f){var _0x40c52e=a0_0x443f,_0xbf433b=_0x4aa430();while(!![]){try{var _0x2a9c0f=parseInt(_0x40c52e(0x2192))/0x1*(-parseInt(_0x4...
L2:
Critical
Browser Credential Phishing
Source appears to collect browser login credentials for exfiltration.
index.jsView on unpkg · L11var a0_0x5a7c88=a0_0x443f;(function(_0x4aa430,_0x26977f){var _0x40c52e=a0_0x443f,_0xbf433b=_0x4aa430();while(!![]){try{var _0x2a9c0f=parseInt(_0x40c52e(0x2192))/0x1*(-parseInt(_0x4...
L2:
Critical
Remote Asset Decode Execute
Source fetches a remote non-code asset, decodes its contents, and dynamically executes the decoded payload.
index.jsView on unpkg · L1Findings
5 Critical5 High4 Medium4 Low
CriticalCommand Output Exfiltrationindex.js
CriticalBrowser Credential Phishingindex.js
CriticalRemote Asset Decode Executeindex.js
CriticalTrigger Reachable Dangerous Capabilityindex.js
CriticalPrevious Version Dangerous Deltaindex.js
HighEvalindex.js
HighSame File Env Network Executionindex.js
HighObfuscated Payload Loaderindex.js
HighTrigger Reachable Command Output Exfiltrationindex.js
HighSemantic Analysis Limitedindex.js
MediumDynamic Requireindex.js
MediumEnvironment Vars
Medium
Affected versions and remediation
This report applies to adbf@1.0.152.
See all versions of adbf for other recorded verdicts.
- Review the evidence and your use of adbf@1.0.152 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.