nixamp@0.67.5
It really whips the terminal's ass. A Winamp-shaped audio player for your terminal.
AI Security Review
scanned 2d ago · by lpm-firewall-aiSource confirms conditional forwarding of the account token to saved remote server URLs, so warn about credential exposure without concluding that the package steals credentials. The account-configured destination and cross-origin product flow make malicious intent unsubstantiated.
Static reason
High-risk behavior combination matched malicious policy.; source matched previously finalized malicious package; routed for review
Trigger
The signed-in radio UI loads or switches to a saved remote server.
Impact
The selected remote server receives the account token, which may expose the associated account session to that server's operator.
Mechanism
The UI obtains the existing account token, places it in the remote server configuration, and adds it as a bearer authorization header on API requests.
AI rationale
Source confirms conditional forwarding of the account token to saved remote server URLs, so warn about credential exposure without concluding that the package steals credentials. The account-configured destination and cross-origin product flow make malicious intent unsubstantiated.
Evidence
web/dist/assets/radio-DmVc3wpH.jsdist/server.js
Network endpoints3
/api/v1/auth/token/api/v1/serversAccount-configured remote server URLs
Decision evidence
public snapshotAI called this Suspicious at 91.0% confidence as Dangerous Capability with medium false-positive risk.
Evidence for warning
- package.json exposes the CLI and includes web/dist in the package files.
- The radio UI retrieves the signed-in account token and reads the account's saved server list.
- The UI attaches that token to requests made to a saved remote server URL.
- The server accepts a caller-supplied URL for an account server, and its source describes carrying the session to another origin.
Evidence against
- The remote destinations come from account server entries, rather than a fixed publisher-controlled host; the server registration route accepts the account holder's URL.
Behavioral surface
ChildProcessCryptoDynamicRequireEnvironmentVarsEvalFilesystemNetwork
HighEntropyStringsMinifiedPossibleObfuscationUrlStrings
Source & flagged code
20 flagged · loading sourcedist/links.jsView file
22import { createHash } from "node:crypto";
L23: import { spawn } from "node:child_process";
L24: import { parseCatalog } from "./catalogs.js";
High
dist/speech.jsView file
230try {
L231: return (await import(__rewriteRelativeImportExtension(name)));
L232: }
Medium
Dynamic Require
Package source references dynamic require/import behavior.
dist/speech.jsView on unpkg · L230dist/catalogs.jsView file
127* decorates the URL on the next line, `#EXTGRP:News` names the group for
L128: * lists that write it that way instead, and `#EXTIMG:https://...` is the
L129: * picture for lists (a podcast's episodes, say) that carry one that way
...
L209: throw new Error(`${source} answered ${answer.status}`);
L210: const text = await answer.text();
L211: if (text.length > maxBytes)
Findings
1 Critical16 High5 Medium6 Low
CriticalHardcoded Runtime Data Exfiltrationweb/dist/assets/radio-DmVc3wpH.js
HighChild Processdist/links.js
HighCommand Output Exfiltrationdist/server.js
HighSandbox Evasion Gated Capabilitydist/manage.js
HighSemantic Analysis Limiteddist/accounts.js#virtual:string-array:round1
HighKnown Malware Source Similaritydist/ai-radio.js
HighKnown Malware Source Similaritydist/session.js
HighKnown Malware Source Similaritysrc/ai-radio.ts
HighKnown Malware Source Similaritysrc/session.ts
HighKnown Malware Source Similaritydist/daemon.js
HighKnown Malware Source Similaritydist/manage.js
Affected versions and remediation
This report applies to nixamp@0.67.5.
See all versions of nixamp for other recorded verdicts.
- Review the evidence and your use of nixamp@0.67.5 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.