registry / nelloz / 0.49.3

nelloz@0.49.3

English-first desktop voice agent with realtime speech, native computer control, secure accounts, memory, and multi-device pairing

Static Scan Results

scanned 2mo ago · by rust-scanner

Static analysis flagged 17 finding(s) at 97.0% confidence. This version is warn-only unless an AI or security-team review confirms malicious behavior.

Static reason

One or more suspicious static signals were detected.; source matched previously finalized malicious package; routed for review

Decision evidence

public snapshot

Behavioral surface

Source

ChildProcessCryptoEnvironmentVarsFilesystemNetworkShellWebSocket

Supply chain

HighEntropyStringsMinifiedUrlStrings

ManifestNo manifest risk signals triggered.

scanned 69 file(s), 1.01 MB of source, external domains: 127.0.0.1, api.deepgram.com, api.duckduckgo.com, api.groq.com, api.openai.com, api.stripe.com, app.daytona.io, duckduckgo.com, example.test, hermes-agent.nousresearch.com, jesucsekcuvtklyidtqn.supabase.co, lite.duckduckgo.com, nelloz.com, nelloz.local

Source & flagged code

9 flagged

scripts/benchmark-observability.mjsView file

Lines 16-56javascript

16function benchmark(name, iterations, callback) {

17 const values = [];

18 for (let index = 0; index < iterations; index += 1) {

19 const startedAt = performance.now();

20 callback(index);

21 values.push(performance.now() - startedAt);

22 }

23 return {

24 name,

25 iterations,

26 p50Ms: Number(percentile(values, 0.5).toFixed(4)),

27 p95Ms: Number(percentile(values, 0.95).toFixed(4)),

28 maxMs: Number(Math.max(...values).toFixed(4)),

29 };

30}

32const home = await fs.mkdtemp(path.join(os.tmpdir(), 'nelloz-observability-benchmark-'));

33const logger = createObservabilityLogger({ home, env: { NELLOZ_LOG_LEVEL: 'debug', NELLOZ_LOG_MAX_BYTES: '16777216' } });

34const privateFixture = {

35 apiKey: 'secret-value-that-must-never-be-written',

36 password: 'password-that-must-never-be-written',

Medium

Secret Pattern

Package contains a possible secret pattern.

scripts/benchmark-observability.mjsView on unpkg · L36

37 userId: 'user-for-benchmark',

38 message: 'private voice content '.repeat(20),

39 latencyMs: 23,

40};

42try {

43 const redaction = benchmark('redact-log-details', 10_000, () => redactLogDetails(privateFixture));

44 const logging = benchmark('enqueue-structured-events', 2_000, (index) => {

45 logger.info('benchmark.event', { ...privateFixture, index }, { traceId: `benchmark-${index}` });

46 });

47 const flushStartedAt = performance.now();

48 await logger.flush();

49 const flushMs = Number((performance.now() - flushStartedAt).toFixed(3));

50 const report = await readObservabilityLogs({ home, lines: 2_100 });

51 const output = {

52 generatedAt: new Date().toISOString(),

53 benchmarks: [redaction, logging],

54 flushMs,

55 recordsWritten: report.records.length,

56 filesWritten: report.files.length,

dist/payload/scripts/deepgram-full-duplex-smoke.mjsView file

Lines 1-23javascript

1import fs from 'node:fs';

2import path from 'node:path';

3import process from 'node:process';

4import { spawnSync } from 'node:child_process'

lib/hermes-runtime.mjsView file

Lines 1-21javascript

1import { spawn, spawnSync } from 'node:child_process';

2import { randomUUID } from 'node:crypto';

3import { promises as fs } from 'node:fs';

4import os from

lib/cli.mjsView file

Lines 1-24javascript

1import { promises as fs } from 'node:fs';

2import os from 'node:os';

3import path from 'node:path';

4import process from 'node:process'

dist/payload/worker.pyView file

•path = dist/payload/worker.py kind = build_helper sizeBytes = 14023 magicHex = [redacted]

Medium

Ships Build Helper

Package ships non-JavaScript build or shell helper files.

dist/payload/worker.pyView on unpkg

lib/auth.mjsView file

•stage = ast_semantic_analysis; reason = ast_alias_growth_limit_exceeded; limitedFiles = 15

High

Semantic Analysis Limited

A bounded semantic-analysis stage reached its safety limit; remaining detectors completed, but this package requires AI review.

lib/auth.mjsView on unpkg

lib/package-runtime.mjsView file

•matchType = normalized_sha256 matchedPackage = nelloz@0.48.5 matchedPath = lib/package-runtime.mjs matchedIdentity = npm:bmVsbG96:0.48.5 similarity = 1.000 summary = normalized source hash matched finalized malicious source

High

Known Malware Source Similarity

Source file is highly similar to a previously finalized malicious package; route for source-aware review.

lib/package-runtime.mjsView on unpkg

Source & flagged code

9 flagged · loading source

scripts/benchmark-observability.mjsView file

36patternName = generic_password severity = medium line = 36

Medium

Secret Pattern

Package contains a possible secret pattern.

scripts/benchmark-observability.mjsView on unpkg · L36

dist/payload/scripts/deepgram-full-duplex-smoke.mjsView file

3import process from 'node:process'; L4: import { spawnSync } from 'node:child_process'; L5: import { fileURLToPath } from 'node:url';

High

Child Process

Package source references child process execution.

dist/payload/scripts/deepgram-full-duplex-smoke.mjsView on unpkg · L3

lib/hermes-runtime.mjsView file

222

Findings

7 High5 Medium5 Low

HighChild Processdist/payload/scripts/deepgram-full-duplex-smoke.mjs

HighShelllib/hermes-runtime.mjs

HighEntrypoint Build Divergencelib/cli.mjs

HighSame File Env Network Executionlib/cli.mjs

HighSandbox Evasion Gated Capabilitylib/hermes-runtime.mjs

HighSemantic Analysis Limitedlib/auth.mjs

HighKnown Malware Source Similaritylib/package-runtime.mjs

MediumSecret Patternscripts/benchmark-observability.mjs

MediumNetwork

MediumEnvironment Vars

MediumShips Build Helperdist/payload/worker.py

MediumStructural Risk Force Deep Review

LowNon Install Lifecycle Scripts

Affected versions and remediation

This report applies to nelloz@0.49.3.

See all versions of nelloz for other recorded verdicts.

Review the evidence and your use of nelloz@0.49.3 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: 2026-09-04 08:36:10.264Z (UTC).

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