@cotal-ai/cli@0.131.5
Cotal mesh CLI: up, join, watch, console, spawn, mint, channels, history, ext.
AI Security Review
scanned 5d ago · by lpm-firewall-aiThe package enables regular runtime telemetry by default, including in non-interactive use where its notice is not displayed. The collection is limited to usage and installation metadata and has an opt-out, so this warrants a warning rather than a block.
Static reason
One or more suspicious static signals were detected.; source matched previously finalized malicious package; routed for review
Trigger
A recognized CLI command runs while telemetry is enabled; missing settings default to regular telemetry.
Impact
The service receives command and flag names, outcomes, error details, shell and terminal metadata, and installation information without an interactive notice in non-TTY use.
Mechanism
The command hook collects usage and installation events and sends them to https://cotal.ai/v1/telemetry. The notice is only printed on an interactive terminal.
AI rationale
The package enables regular runtime telemetry by default, including in non-interactive use where its notice is not displayed. The collection is limited to usage and installation metadata and has an opt-out, so this warrants a warning rather than a block.
Evidence
dist/command.jsdist/telemetry.js.lpm-dependency-evidence-92cd0a6b-1c25-4bc5-a3cd-433810b85a40/1/npm-linux/dist/telemetry.js~/.cotal/telemetry.json
Network endpoints1
cotal.ai/v1/telemetry
Decision evidence
public snapshotAI called this Suspicious at 96.0% confidence as Dangerous Capability with medium false-positive risk.
Evidence for warning
- The CLI arms telemetry for recognized commands before they run.
- When telemetry settings are absent, the workspace dependency creates them at the regular level.
- Regular command events include command names, flags, error details, shell, and terminal metadata.
- Telemetry is sent to https://cotal.ai/v1/telemetry.
Evidence against
- package.json defines no lifecycle scripts.
- Telemetry can be disabled, and its notice is shown on interactive terminals.
- Agent skills are written by the explicit setup flow and limited to Cotal-managed skill files.
- The global npm install is offered during setup and guarded by an interactive prompt when applicable.
Behavioral surface
ChildProcessCryptoDynamicRequireEnvironmentVarsFilesystemNetworkShell
HighEntropyStringsUrlStrings
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/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
HighKnown Malware Source Similaritydist/commands/setup-start.js
Affected versions and remediation
This report applies to @cotal-ai/cli@0.131.5.
See all versions of @cotal-ai/cli for other recorded verdicts.
- Review the evidence and your use of @cotal-ai/cli@0.131.5 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.