@cotal-ai/cli@0.129.1
Cotal mesh CLI: up, join, watch, console, spawn, mint, channels, history, ext.
AI Security Review
scanned 6d ago · by lpm-firewall-aiThe package sends disclosed, usage-focused runtime telemetry to its own endpoint, but the default regular setting can transmit events in non-interactive sessions without displaying the notice. This is a warning-level telemetry concern, not evidence of malicious behavior.
Static reason
One or more suspicious static signals were detected.; source matched previously finalized malicious package; routed for review
Trigger
Running a recognized CLI command when telemetry is not disabled.
Impact
Cotal receives usage events. The stated fields are command names, outcomes, masked error text, version, OS, and installed connectors.
Mechanism
The CLI reads the telemetry level, arms an exit handler, and sends the event to https://cotal.ai/v1/telemetry. A new settings file defaults to regular telemetry; the notice is conditional on a TTY.
AI rationale
The package sends disclosed, usage-focused runtime telemetry to its own endpoint, but the default regular setting can transmit events in non-interactive sessions without displaying the notice. This is a warning-level telemetry concern, not evidence of malicious behavior.
Evidence
dist/telemetry.js.lpm-dependency-evidence-643138d8-04fd-48dc-950f-5c448342090a/1/npm-linux/dist/telemetry.js
Network endpoints1
cotal.ai/v1/telemetry
Decision evidence
public snapshotAI called this Suspicious at 88.0% confidence as Dangerous Capability with medium false-positive risk.
Evidence for warning
- When no telemetry settings file exists, the workspace dependency creates one with the level set to regular.
- The CLI arms telemetry for recognized commands and sends an event on process exit to https://cotal.ai/v1/telemetry.
- The regular-level notice is shown only when stderr is a TTY; telemetry can therefore be sent without that notice in non-interactive use.
- The notice describes the collected fields as command names, outcomes, masked error text, version, OS, and installed connectors.
Evidence against
- The manifest has no install, preinstall, or postinstall lifecycle scripts.
- Cross-vendor agent skills are reconciled by the explicit setup flow, and the installer writes only Cotal-authored skill files under ~/.agents/skills.
Behavioral surface
ChildProcessCryptoDynamicRequireEnvironmentVarsFilesystemNetworkShell
HighEntropyStrings
Source & flagged code
18 flaggeddist/seed/reconcile.jsView file
1import { existsSync, renameSync } from "node:fs";
L2: import { spawnSync } from "node:child_process";
L3: import { randomBytes } from "node:crypto";
High
Child Process
Package source references child process execution.
dist/seed/reconcile.jsView on unpkg · L1dist/lib/persona-files.jsView file
274process.on(signal, pass);
L275: const child = program ? spawn(program, [draft], { stdio: "inherit" }) : spawn(`${editor} ${quoted}`, { shell: true, stdio: "inherit" });
L276: return new Promise(resolve => {
High
dist/lib/nats-bin.jsView file
Findings
17 High4 Medium4 Low
HighChild Processdist/seed/reconcile.js
HighShelldist/lib/persona-files.js
HighTrigger Reachable External Ai Agent Control Surface Mutationdist/lib/agent-skills.js
HighRuntime Package Installdist/commands/setup.js
HighSemantic Analysis Limiteddist/commands/ext.js
HighKnown Malware Source Similaritydist/telemetry.js
HighKnown Malware Source Similaritydist/commands/spawn.js
HighKnown Malware Source Similaritydist/commands/up.js
HighKnown Malware Source Similaritydist/lib/self-exec.js
HighKnown Malware Source Similaritydist/lib/service-process.js
HighKnown Malware Source Similaritydist/commands/ext.js
Affected versions and remediation
This report applies to @cotal-ai/cli@0.129.1.
See all versions of @cotal-ai/cli for other recorded verdicts.
- Review the evidence and your use of @cotal-ai/cli@0.129.1 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: .
Think this verdict is wrong? Dispute it on GitHub or read how reports are produced.