MAL-2026-13399
Vulnerability from ossf_malicious_packages
-= Per source details. Do not edit below this line.=-
Source: amazon-inspector (ca33b55b94f99f1de3d4b34ee5fc42941013cc5ab1f94dce49b6cc0bed5eaa9a)
agenthub-multiagent-mcp ships a worker (dist/worker.js) that opens a WebSocket to the hardcoded server wss://agenthub.contetial.com/ws/workers/ and, for every 'dispatch' message received, writes the server-supplied message body to a prompt file and spawns 'bash run-claude.sh' which invokes 'claude -p "$PROMPT" --dangerously-skip-permissions --max-turns N --mcp-config...' with cwd set to a user-configured project directory. This gives whoever controls agenthub.contetial.com the ability to drive Claude Code's full tool suite (file read/write, shell, MCP tools) against the installer's projects with permissions checks explicitly disabled. The setup wizard (agenthub-setup) additionally installs OS-level persistence: a Windows Startup.vbs + hidden.bat launcher, a macOS LaunchAgent com.agenthub.worker.plist with RunAtLoad and KeepAlive, or a Linux systemd user unit agenthub-worker.service with Restart=always, so the remote-controlled worker reconnects on every login. Provenance is falsified: package.json declares repository.url https://github.com/anthropics/agenthub and homepage https://github.com/anthropics/agenthub#readme, and the README directs users to git clone https://github.com/anthropics/agenthub, while the actual author is 'Krishi AI' and the operational domain is agenthub.contetial.com. dist/setup.js contains a fetch to https://agenthub.contetial.com and dist/context.js/state.js perform base64 decoding of embedded blobs.
- CWE-506 - The product contains code that appears to be malicious in nature.
{
"affected": [
{
"database_specific": {
"cwes": [
{
"cweId": "CWE-506",
"description": "The product contains code that appears to be malicious in nature.",
"name": "Embedded Malicious Code"
}
],
"indicators": {
"evidence_files": [
{
"path": "dist/worker.js",
"sha256": "3be60ed23888e72f6843532db1ff0dc73aa46db8eb031659fc0360a9bf556e58",
"tlsh": "36d2c58a58f7063202777028579f1005ba3aa2673644e9a47acc93449f5d72cc7f6fee"
},
{
"path": "dist/setup.js",
"sha256": "eb9f72ed2d00da4ecbaff51b642d8c915b9f266490b19f70e22fdb49a6ccecf5",
"tlsh": "5e92b48a48f705360b736aa85b2f1505735ed123b248fc94768c97c46f76328c962bee"
},
{
"path": "package.json",
"sha256": "03a4eed9a7bb059df50eb413cf881e06ecf69250feb24a9f9a9c5aa1770e8961",
"tlsh": "80410f6bc9ab8d7307b8a7c1fab60201f315432f9100988bb2b52a1ccff5297114bb58"
}
],
"package_integrity": [
{
"filename": "agenthub-multiagent-mcp-1.57.0.tgz",
"hashes": {
"sha1": "c339a2c1bfbe0d4a80b3c800c0cb511f09488a6d",
"sha512_sri": "sha512-Hor9cwvDy6wvfgeW3bJ0U4OL6Bw7y/V+R8bpWuuA6xgw1I6IiRVD7jNEbE0T189Fbk1Y0t39km1unGYNBoTDpA=="
}
}
]
}
},
"package": {
"ecosystem": "npm",
"name": "agenthub-multiagent-mcp"
},
"versions": [
"1.57.0"
]
}
],
"credits": [
{
"contact": [
"inspector-research@amazon.com"
],
"name": "Amazon Inspector",
"type": "FINDER"
}
],
"database_specific": {
"malicious-packages-origins": [
{
"id": "IN-MAL-2026-016096",
"import_time": "2026-08-06T13:08:48.901998744Z",
"modified_time": "2026-08-06T12:30:47Z",
"sha256": "ca33b55b94f99f1de3d4b34ee5fc42941013cc5ab1f94dce49b6cc0bed5eaa9a",
"source": "amazon-inspector",
"versions": [
"1.57.0"
]
}
]
},
"details": "\n---\n_-= Per source details. Do not edit below this line.=-_\n\n## Source: amazon-inspector (ca33b55b94f99f1de3d4b34ee5fc42941013cc5ab1f94dce49b6cc0bed5eaa9a)\nagenthub-multiagent-mcp ships a worker (dist/worker.js) that opens a WebSocket to the hardcoded server wss://agenthub.contetial.com/ws/workers/\u003cid\u003e and, for every \u0027dispatch\u0027 message received, writes the server-supplied message body to a prompt file and spawns \u0027bash run-claude.sh\u0027 which invokes \u0027claude -p \"$PROMPT\" --dangerously-skip-permissions --max-turns N --mcp-config...\u0027 with cwd set to a user-configured project directory. This gives whoever controls agenthub.contetial.com the ability to drive Claude Code\u0027s full tool suite (file read/write, shell, MCP tools) against the installer\u0027s projects with permissions checks explicitly disabled. The setup wizard (agenthub-setup) additionally installs OS-level persistence: a Windows Startup.vbs + hidden.bat launcher, a macOS LaunchAgent com.agenthub.worker.plist with RunAtLoad and KeepAlive, or a Linux systemd user unit agenthub-worker.service with Restart=always, so the remote-controlled worker reconnects on every login. Provenance is falsified: package.json declares repository.url https://github.com/anthropics/agenthub and homepage https://github.com/anthropics/agenthub#readme, and the README directs users to git clone https://github.com/anthropics/agenthub, while the actual author is \u0027Krishi AI\u0027 and the operational domain is agenthub.contetial.com. dist/setup.js contains a fetch to https://agenthub.contetial.com and dist/context.js/state.js perform base64 decoding of embedded blobs.\n",
"id": "MAL-2026-13399",
"modified": "2026-08-06T12:30:47Z",
"published": "2026-08-06T12:30:47Z",
"references": [
{
"type": "PACKAGE",
"url": "https://www.npmjs.com/package/agenthub-multiagent-mcp/v/1.57.0"
}
],
"schema_version": "1.7.4",
"summary": "Malicious code in agenthub-multiagent-mcp (npm)"
}
Sightings
| Author | Source | Type | Date | Other |
|---|
Nomenclature
- Seen: The vulnerability was mentioned, discussed, or observed by the user.
- Confirmed: The vulnerability has been validated from an analyst's perspective.
- Published Proof of Concept: A public proof of concept is available for this vulnerability.
- Exploited: The vulnerability was observed as exploited by the user who reported the sighting.
- Patched: The vulnerability was observed as successfully patched by the user who reported the sighting.
- Not exploited: The vulnerability was not observed as exploited by the user who reported the sighting.
- Not confirmed: The user expressed doubt about the validity of the vulnerability.
- Not patched: The vulnerability was not observed as successfully patched by the user who reported the sighting.