registry / @cotal-ai/cli / 0.133.10
@cotal-ai/cli@0.133.10
Cotal mesh CLI: up, join, watch, console, spawn, mint, channels, history, ext.
AI Security Review
scanned 4d ago · by lpm-firewall-aiThe suspicious scanner hits resolve to ordinary CLI behavior: connector seeding is gated behind a validated invocation, skill writes occur through setup, and feedback transmission is user initiated. No install-time exfiltration, hidden account action, or unrelated agent-control mutation was found.
Static reason
One or more suspicious static signals were detected.
Trigger
Explicit CLI commands; connector seeding follows a validated command invocation.
Impact
No unauthorized credential disclosure, account action, or unrelated agent configuration change was found.
Mechanism
Setup reconciles the package’s authored skills, and connector reconciliation stages built-in payloads. Feedback and mesh connections use user-provided or user-selected inputs.
AI rationale
The suspicious scanner hits resolve to ordinary CLI behavior: connector seeding is gated behind a validated invocation, skill writes occur through setup, and feedback transmission is user initiated. No install-time exfiltration, hidden account action, or unrelated agent-control mutation was found.
Evidence
~/.agents/skills/<skill>/SKILL.md
Network endpoints3
broker.cotal.ai/v1/feedbackcotal.ai/v1/feedbackUser-selected mesh endpoints
Decision evidence
public snapshotAI called this Clean at 95.0% confidence as Benign with low false-positive risk.
Evidence for AI clean decision
- The package exposes a JavaScript CLI entry and declares no install lifecycle hook.
- Connector reconciliation runs after a validated CLI invocation; it stages a built-in payload and invokes this CLI’s `ext add` command for it.
- Agent skills are reconciled through `cotal setup`; the code writes only the shipped `SKILL.md` files under `~/.agents/skills` and preserves divergent copies.
- Feedback is submitted by an explicit command and sends the provided summary and optional details to the configured or Cotal intake.
- `join` uses a link or credentials supplied for the selected mesh, or resolves the user’s running mesh when invoked.
Evidence against
- The manifest’s executable behavior is exposed through the CLI entry point, with no install hook declared.
- The flagged shell use launches the configured editor for persona editing.
- The skills write is part of explicit setup and is limited to the package’s authored skill files.
Behavioral surface
ChildProcessCryptoDynamicRequireEnvironmentVars
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@0.133.10.
See all versions of @cotal-ai/cli for other recorded verdicts.
- Review the evidence and your use of @cotal-ai/cli@0.133.10 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.