@cotal-ai/cli@0.133.0
Cotal mesh CLI: up, join, watch, console, spawn, mint, channels, history, ext.
AI Security Review
scanned 4d ago · by lpm-firewall-aiThis is warning-level runtime telemetry: regular collection starts by default, and noninteractive runs can send events without the one-time notice. The destination is the package’s own service and the inspected payload does not contain credentials or command contents, so this is not malicious.
Static reason
One or more suspicious static signals were detected.
Trigger
A normal CLI command when telemetry is enabled; a missing settings file defaults to regular telemetry.
Impact
Usage and install metadata is reported without a visible notice in noninteractive runs; no credential or content exfiltration was identified.
Mechanism
The CLI queues command and periodic install events for a POST to Cotal’s telemetry endpoint at process exit. Its notice is gated on stderr being a TTY.
AI rationale
This is warning-level runtime telemetry: regular collection starts by default, and noninteractive runs can send events without the one-time notice. The destination is the package’s own service and the inspected payload does not contain credentials or command contents, so this is not malicious.
Evidence
dist/telemetry.jsdist/command.js.lpm-dependency-evidence-f1a42b6f-b3f5-45da-973b-c44c9adb83b8/1/npm-linux/dist/telemetry.jstelemetry.json under the user's Cotal directory
Network endpoints1
cotal.ai/v1/telemetry
Decision evidence
public snapshotAI called this Suspicious at 88.0% confidence as Unknown with medium false-positive risk.
Evidence for warning
- A missing telemetry settings file is created with level `regular` and a random install ID; the CLI arms telemetry for commands.
- The one-time regular-telemetry notice is printed only when stderr is a TTY, so noninteractive runs can send telemetry without that notice.
- The CLI sends telemetry on process exit to `https://cotal.ai/v1/telemetry`.
- Regular command events include command names, outcomes, flag names, a hashed account identifier, and scrubbed error details; a daily install event reports connector and extension metadata.
Evidence against
- The telemetry notice describes the collection and points users to `cotal telemetry off`; the package also offers minimal and regular levels.
- The inspected events do not include command argument values or file contents, and the destination is Cotal’s own service.
- The manifest declares no install lifecycle hooks; flagged skill writes and global npm installation are behind an explicit `cotal setup` flow.
Behavioral surface
ChildProcessCryptoDynamicRequire
Source & flagged code
6 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
5 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
MediumDynamic Requiredist/lib/nats-bin.js
MediumNetwork
MediumEnvironment Vars
MediumStructural Risk Force Deep Review
LowScripts Present
LowFilesystem
LowHigh Entropy Strings
LowUrl Strings
Affected versions and remediation
This report applies to @cotal-ai/cli@0.133.0.
See all versions of @cotal-ai/cli for other recorded verdicts.
- Review the evidence and your use of @cotal-ai/cli@0.133.0 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.