Loading npm security reports…

claude-token-saver@3.35.1: AI-agent control-surface risk | LPM Firewall

registry / claude-token-saver / 3.35.1

claude-token-saver@3.35.1

Route the easy work your expensive Claude model keeps repeating down to haiku/sonnet — post-hoc session analysis, no realtime router, no extra LLM calls.

AI Security Review

scanned 27d ago · by lpm-firewall-ai

LPM flags this version as an AI-agent control-surface risk. Installing the package automatically configures Claude Code to run this package at session start and on every submitted prompt. It also adds global Claude instructions without an interactive consent step during postinstall.

Static reason

High-risk behavior combination matched malicious policy.; source matched previously finalized malicious package; routed for review; source fingerprint signature matched known malicious package; routed for review

Trigger

npm postinstall runs `node bin/cli.js install`.

Impact

The package gains recurring execution and can inject package-controlled behavior into future Claude Code sessions.

Mechanism

Automatic mutation of Claude Code hooks, status-line settings, skills, and global instructions.

Policy narrative

On installation, the postinstall lifecycle invokes the package installer. In unattended contexts, including postinstall, it proceeds with automatic defaults. The installer writes Claude Code settings that register package commands for SessionStart and UserPromptSubmit, configures a status-line command, installs a package-controlled skill, and can append global instructions to ~/.claude/CLAUDE.md. This creates persistent package-controlled execution and prompt-control behavior across later Claude Code sessions without explicit consent at install time.

AI rationale

This is an automatic postinstall mutation of a broad AI-agent control surface, not an explicit user-command setup. The recurring hooks and global instruction writes establish concrete persistent control of future agent sessions.

Evidence

package.jsonbin/cli.jssrc/commands/install.jssrc/installer.jssrc/harness.js~/.claude/settings.json

Source & flagged code

13 flagged · loading source

package.jsonView file

•scripts.postinstall = node bin/cli.js install || true

High

Install Time Lifecycle Scripts

Package defines install-time lifecycle scripts.

package.jsonView on unpkg

•scripts.postinstall = node bin/cli.js install || true

Medium

Ambiguous Install Lifecycle Script

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

package.jsonView on unpkg

src/doc2md.cjsView file

•matchType = normalized_sha256 matchedPackage = claude-token-saver@3.35.0 matchedPath = src/doc2md.cjs matchedIdentity = npm:Y2xhdWRlLXRva2VuLXNhdmVy:3.35.0 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.

src/doc2md.cjsView on unpkg

•matchType = malicious_source_fingerprint_signature signature = 27de6161b10700e3 signatureType = suspicious_hashes sourceLabel = final_verdict:malicious matchedPackage = claude-token-saver@3.35.0 matchedPath = src/doc2md.cjs matchedIdentity = npm:Y2xhdWRlLXRva2VuLXNhdmVy:3.35.0 similarity = 1.000 shingleOverlap = 43 summary = package final verdict is malicious

High

Known Malware Source Fingerprint Signature

Source fingerprint signature matches a known malicious package signature; route for source-aware review.

src/doc2md.cjsView on unpkg

24L25: const fs = require('node:fs'); L26: const os = require('node:os');

Medium

Dynamic Require

Package source references dynamic require/import behavior.

src/doc2md.cjsView on unpkg · L24

src/seed-rules.jsView file

53try { L54: const data = JSON.parse(readFileSync(seedStatePath(), 'utf8')); L55: return data && typeof data.decided === 'object' && data.decided ? data : { decided: {} };

Low

Weak Crypto

Package source references weak cryptographic algorithms.

src/seed-rules.jsView on unpkg · L53

bin/cli.jsView file

10Install-time code directly mutates a foreign AI-agent control surface: L104: claude-token-saver uninstall remove everything install added L105: claude-token-saver harness check score the harness setup in CLAUDE.md L106: claude-token-saver harness analyze run the harness transcript analysis manually ... L234: L235: // Subcommand: harness — manage the project's CLAUDE.md harness rules. L236: // claude-token-saver harness init # write CLAUDE.md (5 sections) + ratchet.md L237: // claude-token-saver harness uninit # remove harness block from CLAUDE.md (backup kept) L238: // claude-token-saver harness check # show 🅷 N/5 + which sections are missing ... L249: // claude-token-saver compact-window # status L250: // claude-token-saver comp

Critical

Ai Agent Control Hijack

Source creates an unconsented AI-agent control surface through install-time mutation or a default unauthenticated remote skill channel.

bin/cli.jsView on unpkg · L10

examples/statusline-command.ps1View file

•path = examples/statusline-command.ps1 kind = build_helper sizeBytes = 1619 magicHex = [redacted]

Medium

Ships Build Helper

Package ships non-JavaScript build or shell helper files.

examples/statusline-command.ps1View on unpkg

src/advice.js#virtual:normalized:round1View file

•stage = ast_semantic_analysis; reason = ast_parse_error; limitedFiles = 4

High

Semantic Analysis Limited

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

src/advice.js#virtual:normalized:round1View on unpkg

src/korean-style.jsView file

•matchType = normalized_sha256 matchedPackage = claude-token-saver@3.35.0 matchedPath = src/korean-style.js matchedIdentity = npm:Y2xhdWRlLXRva2VuLXNhdmVy:3.35.0 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.

src/korean-style.jsView on unpkg

src/litellm-budget.jsView file

•matchType = normalized_sha256 matchedPackage = claude-token-saver@3.35.0 matchedPath = src/litellm-budget.js matchedIdentity = npm:Y2xhdWRlLXRva2VuLXNhdmVy:3.35.0 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.

src/litellm-budget.jsView on unpkg

src/update-check.jsView file

•matchType = normalized_sha256 matchedPackage = claude-token-saver@3.35.0 matchedPath = src/update-check.js matchedIdentity = npm:Y2xhdWRlLXRva2VuLXNhdmVy:3.35.0 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.

src/update-check.jsView on unpkg

src/model-alias.jsView file

•matchType = normalized_sha256 matchedPackage = claude-token-saver@3.35.0 matchedPath = src/model-alias.js matchedIdentity = npm:Y2xhdWRlLXRva2VuLXNhdmVy:3.35.0 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.

src/model-alias.jsView on unpkg

Source & flagged code

13 flagged

package.jsonView file

Lines 1-32json

1{

2 "name": "claude-token-saver",

3 "version": "3.35.1",

4 "description": "Route the easy work your expensive Claude model keeps repeating down to haiku/sonnet — post-hoc session analysis, no realtime router, no extra LLM calls.",

5 "type": "module",

6 "bin": {

7 "claude-token-saver": "bin/cli.js"

8 },

9 "scripts": {

10 "test": "node --test",

11 "docs:statusline": "node scripts/docs-statusline.mjs",

12 "postinstall": "node bin/cli.js install || true"

High

Install Time Lifecycle Scripts

Package defines install-time lifecycle scripts.

package.jsonView on unpkg · L12

Medium

Ambiguous Install Lifecycle Script

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

package.jsonView on unpkg · L12

13 },

14 "files": [

15 "bin/",

16 "src/",

17 "presets/",

18 "examples/",

19 "README.md",

20 "README.en.md",

21 "LICENSE"

22 ],

23 "engines": {

24 "node": ">=18"

25 },

26 "keywords": [

27 "claude",

28 "claude-code",

29 "anthropic",

30 "model-routing",

31 "llm-router",

32 "delegation",

src/doc2md.cjsView file

Lines 4-44javascript

4 *

5 * A pptx or xlsx read straight into the context window is close to the worst

6 * thing a token-saving tool can allow: the bytes are unreadable to the model,

7 * so it either gets nothing useful or spends a fortune finding that out. This

8 * intercepts the Read, converts the file once, caches the result, and points

src/seed-rules.jsView file

Lines 33-73javascript

33 addModelRule,

34 composeRuleText,

35 loadModelRules,

36 modelRuleBaseText,

37 syncAllFiles,

38} from './model-rules.js';

39import {

40 findProjectRoot,

41 harnessListRules,

42 harnessPromote,

bin/cli.jsView file

Lines 1-30javascript

1#!/usr/bin/env node

2

3/**

4 * claude-token-saver CLI (formerly claude-cache-monitor)

5 *

6 * Usage:

7 * npx claude-token-saver # default report (last 30 days)

8 * npx claude-token-saver --days 7 # last 7 days

9 * npx claude-token-saver --format json # JSON output

10 * npx claude-token-saver --format csv # CSV output

11 * npx claude-token-saver --project myproj # filter by project

examples/statusline-command.ps1View file

•path = examples/statusline-command.ps1 kind = build_helper sizeBytes = 1619 magicHex = [redacted]

Medium

Ships Build Helper

Package ships non-JavaScript build or shell helper files.

examples/statusline-command.ps1View on unpkg

src/advice.js#virtual:normalized:round1View file

•stage = ast_semantic_analysis; reason = ast_parse_error; limitedFiles = 4

High

Semantic Analysis Limited

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

src/advice.js#virtual:normalized:round1View on unpkg

src/korean-style.jsView file

•matchType = normalized_sha256 matchedPackage = claude-token-saver@3.35.0 matchedPath = src/korean-style.js matchedIdentity = npm:Y2xhdWRlLXRva2VuLXNhdmVy:3.35.0 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.

src/korean-style.jsView on unpkg

src/litellm-budget.jsView file

•matchType = normalized_sha256 matchedPackage = claude-token-saver@3.35.0 matchedPath = src/litellm-budget.js matchedIdentity = npm:Y2xhdWRlLXRva2VuLXNhdmVy:3.35.0 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.

src/litellm-budget.jsView on unpkg

src/update-check.jsView file

•matchType = normalized_sha256 matchedPackage = claude-token-saver@3.35.0 matchedPath = src/update-check.js matchedIdentity = npm:Y2xhdWRlLXRva2VuLXNhdmVy:3.35.0 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.

src/update-check.jsView on unpkg

src/model-alias.jsView file

•matchType = normalized_sha256 matchedPackage = claude-token-saver@3.35.0 matchedPath = src/model-alias.js matchedIdentity = npm:Y2xhdWRlLXRva2VuLXNhdmVy:3.35.0 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.

src/model-alias.jsView on unpkg