Lightcone Session

ReviewAudited by ClawScan on May 10, 2026.

Overview

This skill is mostly coherent for cloud browser/desktop control, but it exposes broad session automation plus a raw shell-command debug action without clear guardrails.

Review this skill before installing if you plan to use it for logged-in sites or sensitive workflows. It appears purpose-aligned, but you should explicitly approve credential entry, account-changing actions, and any use of the debug shell command, and always close Lightcone sessions when finished.

Findings (3)

Artifact-based informational review of SKILL.md, metadata, install specs, static scan signals, and capability signals. ClawScan does not execute the skill or run runtime probes.

Concern

ASI02: Tool Misuse and Exploitation

What this means

If used carelessly, the agent could run arbitrary commands inside the cloud session or combine shell access with logged-in browser state.

Why it was flagged

The skill documents an escape-hatch debug action that runs shell commands in the cloud computer, but does not state limits or approval requirements for when the agent may use it.

Skill content

`command` | For debug | Shell command to run ... `debug` | Command output | Run shell command in the computer

Recommendation

Use the debug action only when explicitly needed, and require user confirmation before shell commands, account-changing actions, or actions involving sensitive sessions.

Note

ASI03: Identity and Privilege Abuse

What this means

The skill can act through the Lightcone account associated with the provided API key.

Why it was flagged

The skill requires a Lightcone/Tzafon API key, which is expected for creating and controlling cloud sessions, and it is disclosed in the metadata.

Skill content

`requires":{"env":["TZAFON_API_KEY"]},"primaryEnv":"TZAFON_API_KEY"

Recommendation

Use a dedicated or least-privileged API key if available, and revoke it if you stop using the skill.

Note

ASI07: Insecure Inter-Agent Communication

What this means

Credentials or private website content entered into the cloud session may be visible to or processed by the external session provider.

Why it was flagged

The documented workflow can send credentials, screenshots, and page HTML through a cloud-computer provider session. This is aligned with the purpose, but sensitive data boundaries are not explained in the artifact.

Skill content

Activate when user needs ... login sequences ... `type`, text: `password123` ... `html` | Page HTML | Extract page source

Recommendation

Avoid entering highly sensitive credentials unless you trust the provider and understand its data handling; close sessions when finished.