claude-multiacc@2.0.55

Unified Claude Code and OpenAI Codex subscription pooling with quota-aware selection.

AI Security Review

scanned 12d ago · by lpm-firewall-ai

The install hook can automatically modify shell command resolution and scheduled tasks during global installation, creating an agent integration lifecycle risk. The behavior appears tied to the package's account-management purpose, so the evidence supports a warning rather than a block.

Static reason

One or more suspicious static signals were detected.

Trigger

A global npm install or update on a supported non-CI platform, unless the postinstall skip flag is set.

Impact

This changes command resolution and creates a recurring task on the installing machine; the inspected evidence does not establish credential theft or other malicious behavior.

Mechanism

The postinstall launches install.sh, which configures the package shim and scheduled account checks.

AI rationale

Evidence

package.jsonscripts/postinstall.mjsinstall.shlib/install_actions.shshell startup configurationuser crontab

Decision evidence

public snapshot

AI called this Suspicious at 88.0% confidence as Dangerous Capability with medium false-positive risk.

Evidence for warning

The package runs a postinstall hook that invokes its installer during eligible global installs.
The installer moves the package shim to the front of PATH, affecting which Claude command launches.
The installer adds a recurring account-limit check to the user's crontab.
The flagged test file contains unit tests for transcript classification; the inspected excerpt does not show a hidden payload.
The account manager refreshes credentials through Anthropic's OAuth token endpoint.

Evidence against

The postinstall skips local installs, CI, Windows, and installs with its skip flag set.
The recurring setup is part of the package's own account-management installer, and the inspected source shows no unrelated endpoint or payload execution.

Behavioral surface

Source

ChildProcessEnvironmentVarsFilesystem

Supply chain

HighEntropyStrings

ManifestNo manifest risk signals triggered.

scanned 2 file(s), 6.59 KB of source

Source & flagged code

7 flagged

package.jsonView file

Lines 36-69json

36 "gowalk"

37 ],

38 "author": {

39 "name": "Gowalk"

40 },

41 "license": "MIT",

42 "repository": {

43 "type": "git",

44 "url": "git+https://github.com/gowalk-public/claude-multiacc.git"

45 },

46 "homepage": "https://github.com/gowalk-public/claude-multiacc#readme",

47 "bugs": {

48 "url": "https://github.com/gowalk-public/claude-multiacc/issues"

49 },

50 "publishConfig": {

51 "access": "public"

52 },

53 "engines": {

54 "node": ">=18"

55 },

56 "scripts": {

57 "postinstall": "node scripts/postinstall.mjs",

High

Install Time Lifecycle Scripts

Package defines install-time lifecycle scripts.

package.jsonView on unpkg · L57

Medium

Ambiguous Install Lifecycle Script

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

package.jsonView on unpkg · L57

58 "test": "npm run test:core && npm run test:seeding && npm run test:commands",

Medium

Suspicious Lifecycle Evidence

The package runs a postinstall hook that invokes its installer during eligible global installs.

package.jsonView on unpkg · L56

59 "test:core": "bash tests/run-tests.sh && npm run test:selector && python3 tests/test_shim_path.py",

60 "test:selector": "python3 -m unittest discover -s tests -p 'test_selector*.py'",

61 "test:seeding": "python3 tests/test_account_seeding.py && python3 tests/test_claude_pool_prepare.py && python3 tests/test_mcp_registry.py && npm run test:codex-settings",

62 "test:codex-settings": "python3 tests/test_codex_config_edit.py && python3 tests/test_codex_settings.py",

63 "test:commands": "python3 tests/test_packaged_commands.py"

64 },

65 "dependencies": {

66 "update-notifier": "^7.3.1"

67 }

68}

install.shView file

•path = install.sh kind = build_helper sizeBytes = 10485 magicHex = [redacted]

Medium

Ships Build Helper

Package ships non-JavaScript build or shell helper files.

install.shView on unpkg

tests/test_autoresume.pyView file

Lines 235-275text

235 def test_server_side_errors_are_transient(self):

236 for code in ("overloaded", "server_error", "unknown"):

237 self.assertEqual(self.verdict(cc_api_error(self.T, code))["class"], "transient")

238

239 def test_auth_and_blocked(self):

240 self.assertEqual(self.verdict(cc_api_error(

241 self.T, "authentication_failed", "Not logged in · Please run /login"))["class"],

242 "auth")

243 for code in ("oauth_org_not_allowed", "account_on_hold", "billing_error",

244 "verification_required"):

245 verdict = self.verdict(cc_api_error(self.T, code))

bin/claude-accountsView file

•stage = ast_semantic_analysis; reason = ast_parse_error; limitedFiles = 3

High

Semantic Analysis Limited

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

bin/claude-accountsView on unpkg

Source & flagged code

7 flagged · loading source

package.jsonView file

•scripts.postinstall = node scripts/postinstall.mjs

High

Install Time Lifecycle Scripts

Package defines install-time lifecycle scripts.

package.jsonView on unpkg

•scripts.postinstall = node scripts/postinstall.mjs

Medium

Ambiguous Install Lifecycle Script

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

package.jsonView on unpkg

56"scripts": { "postinstall": "node scripts/postinstall.mjs", "test": "npm run test:core && npm run test:seeding && npm ru

Medium

Suspicious Lifecycle Evidence

The package runs a postinstall hook that invokes its installer during eligible global installs.

package.jsonView on unpkg · L56

install.shView file

Findings

3 High6 Medium3 Low

HighInstall Time Lifecycle Scriptspackage.json

HighPayload In Excluded Dirtests/test_autoresume.py

HighSemantic Analysis Limitedbin/claude-accounts

MediumAmbiguous Install Lifecycle Scriptpackage.json

MediumEnvironment Vars

MediumShips Build Helperinstall.sh

MediumStructural Risk Force Deep Review

MediumSuspicious Lifecycle Evidencepackage.json

MediumAi Review Evidencetests/test_autoresume.py

LowScripts Present

LowFilesystem

LowHigh Entropy Strings

Affected versions and remediation

This report applies to claude-multiacc@2.0.55.

See all versions of claude-multiacc for other recorded verdicts.

Review the evidence and your use of claude-multiacc@2.0.55 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: 2026-10-10 21:31:16.790Z (UTC).

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

AI Security Review

Decision evidence

Source & flagged code

Source & flagged code

Findings

Affected versions and remediation

Runtime surface

Behavioral surface

Source & flagged code