Loading npm security reports…
Loading npm security reports…
@monoes/monomindcli@2.24.9: Clean npm security report (Allow) | LPM Firewall
npm / @monoes/monomindcli / 2.24.9
@monoes/monomindcli@2.24.9 CLI engine for Monomind — an open-source MCP server for Claude Code, OpenCode, Antigravity, Kimi Code, and Codex, with a codebase knowledge graph (tree-sitter + SQLite), persistent memory, multi-agent task coordination, and session hooks. Apache 2.0 licen
Allowed — no malicious behavior detected
19 low-signal patterns reviewed and cleared.
AI security reviewscanned on Oct 16, 2026, 11:46 UTC · by lpm-firewall-ai The inspected entrypoints and flagged source locations show product-aligned CLI, MCP, and local agent integration behavior, with no install hook or concrete attack behavior established. Scanner labels alone do not support a malicious verdict.
Static reason
High-risk behavior combination matched malicious policy.; previous stored version diff introduced dangerous source
Trigger
Normal use of the CLI, MCP server, or Claude helper scripts.
Impact
No concrete malicious effect or confirmed attack surface was established.
Mechanism
The inspected code reads local agent definitions, settings, and session files, and performs its documented helper and process-management functions.
Decision evidencepublic snapshot Evidence for AI clean decision
package.json exposes CLI and MCP entrypoints; its scripts show build, test, and release tasks, with no install lifecycle hook.
Agent registry and skill helper locations read project/user agent directories and project command Markdown for normal registry discovery.
Hook handlers and utilities read Claude project JSONL and settings files for capture, restore, task, statusline, and token tracking.
MCP server process inspection checks that a PID belongs to a Monomind MCP process; design detection resolves the bundled Monodesign CLI.
Generated pre-commit hook runs staged-file validation and optional project tests; managed Chrome downloads a browser only when requested and missing.
Agent exec requires caller-supplied runtime and prompt input; README describes an MCP server integrating with coding-agent runtimes. Behavioral surface
Source Child process · Crypto · Dynamic require · Environment variables · Eval · Filesystem · Network · Shell · Web socket
Supply chain High entropy strings · URL strings
Manifest No manifest risk signals triggered.
Scanned 1,103 files · 8.02 MB of source
39 external domains referenced
127.0.0.1 accounts.google.com aider.chat antigravity.google api.anthropic.com api.example.com api.github.com api.osv.dev api.z.ai cdnjs.cloudflare.com cli.github.com code.claude.com code.visualstudio.com cursor.com docs.factory.ai docs.github.com docs.openclaw.ai docs.x.ai example.com gemini.google.com git-scm.com github.com hermes-agent.nousresearch.com huggingface.co json-schema.org kiro.dev learn.chatgpt.com monoes.me nodejs.org opencode.ai pi.dev raw.githubusercontent.com registry.npmjs.org services.nvd.nist.gov www.apple.com www.instagram.com www.kimi.com www.linkedin.com x.com Findings3 critical · 5 high · 6 medium · 5 low Source & flagged code11 flagged The lines the automated scan flagged. The review checked them and cleared this version.
High AI reviewer manipulationCleared
Package text addresses the security reviewer or scanner and tries to influence the review outcome.
README.md View on unpkg Scanner record · 1 line isSafe('Ignore all previous instructions'); // → false (~0.04ms)
dist/src/init/codex-generator.js#virtual:normalized:round1 View file Medium Dynamic requireCleared
Package source references dynamic require/import behavior.
dist/src/init/codex-generator.js#virtual:normalized:round1 View on unpkg · L37 Code excerpt · 3 lines 37 }
38 /** Generate the project-local adapter used by Codex"s native hook runner. */\nexport function generateCodexHookScript() {\n return `// Monomind native Codex hook bridge (genera...
39 }
dist/src/mcp-server.js#virtual:normalized:round1 View file Critical Persistence backdoorCleared
Source writes persistence or remote-access backdoor material.
dist/src/mcp-server.js#virtual:normalized:round1 View on unpkg · L18 Critical Trigger reachable dangerous capabilityCleared
A package entrypoint or install-time lifecycle script reaches a source file with blocking dangerous behavior.
dist/src/mcp-server.js#virtual:normalized:round1 View on unpkg · L18 High Trigger reachable persistenceCleared
A manifest entrypoint or package-local install chain reaches persistence behavior.
dist/src/mcp-server.js#virtual:normalized:round1 View on unpkg · L18 Shared code excerpt · 3 findings · 12 lines 18 */
19 import { execSync } from "node:child_process";
20 import { randomUUID } from "node:crypto";
⋯
High Copied package dependency bridgeCleared
Package metadata claims a different repository identity while copied source loads a runtime dependency bridge.
dist/src/commands/design-detect.js View on unpkg · L66 Code excerpt · 4 lines package = @monoes/monomindcli; repositoryIdentity = monomind; dependency = @monoes/monobrowse
66 export async function urlScanEnv () {
67 const { findChrome } = await import ('@monoes/monobrowse' );
68 try {
dist/src/init/helpers-core-scripts.js View file High External AI agent control surface mutationCleared
Runtime or CLI source writes behavior-bearing configuration into a user or project AI-agent control surface.
dist/src/init/helpers-core-scripts.js View on unpkg Scanner record · 29 lines Runtime or CLI code links a write operation to an explicit external AI-agent control path:
ession.metrics)}\`);
return session;
},
update: (key, value) => {
if (!fs.existssync(session_file)) {
console.log('no active session');
return null;
}
const session = json.parse(fs.readfilesync(session_file, 'utf-8'));
session.context[key] = value;
session.updatedat = new date().toisostring();
fs.writefilesync(session_file, json.stringify(session, null, 2));
dist/src/orgrt/aider/monomind_aider_shim.py View file Medium Ships build helperCleared
Package ships non-JavaScript build or shell helper files.
dist/src/orgrt/aider/monomind_aider_shim.py View on unpkg Scanner record · 4 lines path = dist/src/orgrt/aider/monomind_aider_shim.py
kind = build_helper
sizeBytes = 19323
magicHex = [redacted]
dist/src/browser/managed-chrome.js#virtual:normalized:round1 View file High Semantic analysis limitedCleared
A bounded semantic-analysis stage reached its safety limit; remaining detectors completed, but this package requires AI review.
dist/src/browser/managed-chrome.js#virtual:normalized:round1 View on unpkg Scanner record · 1 line stage = ast_semantic_analysis; reason = ast_parse_error; limitedFiles = 16
Critical Previous version dangerous deltaCleared
This package version adds a dangerous source file absent from the previous stored version; route for source-aware review.
dist/src/commands/agent-exec.js View on unpkg Scanner record · 5 lines matchType = previous_version_dangerous_delta
matchedPackage = @monoes/monomindcli@2.24.8
matchedIdentity = npm:QG1vbm9lcy9tb25vbWluZGNsaQ:2.24.8
similarity = 0.983
summary = stored previous version shares package body but lacks this dangerous source file
↓
Cleared
Package metadata 8 findings
Medium Network Cleared
Medium Environment variables Cleared
Medium Structural risk force deep review Cleared
Low Scripts present Cleared
Low High entropy strings Cleared
22 import * as fs from "node:fs";
23 import * as http from "node:http";
24 import * as path from "node:path";
57 import { trackRequest } from "./mcp-tools/request-tracker.js";
58 // ESM-compatible __dirname
59 const __filename = fileURLToPath(import.meta.url);
67 * attackable (a local attacker pre-creates /tmp/monomind-mcp.pid as a symlink
68 * to e.g. ~/.ssh/authorized_keys, then a writeFile clobbers the target).
if (!fs.existssync(session_file)) {
const session = json.parse(fs.readfilesync(session_file, 'utf-8'));
if (session.metrics[name] !== undefined) {
fs.writefilesync(session_file,