Skillv1.0.0

ClawScan security

Panel Generator · ClawHub's context-aware review of the artifact, metadata, and declared behavior.

Scanner verdict

SuspiciousApr 17, 2026, 6:24 PM

Verdict: suspicious
Confidence: medium
Model: gpt-5-mini
Summary: The skill's declared purpose (generate panel-layout videos) matches most of its instructions, but there are several inconsistencies and data-handling behaviors (automatic backend connection, anonymous token creation/storage, and an undocumented config path) that deserve user review before installing.
Guidance: Before installing, confirm how NEMO_TOKEN and session_id will be stored (where on disk, encrypted or not) and whether the skill will ever auto-upload user files without an explicit user action. Ask the publisher for a privacy/data-retention policy and for clarification about the config path (~/.config/nemovideo/) and why the registry metadata omitted it. If you want to limit risk, require explicit user consent before the skill contacts the external API or generates/stores an anonymous token, and avoid putting any sensitive images or private media into the skill until you trust the backend.

Review Dimensions

Purpose & Capability: okThe skill is an instruction-only panel/video generator and requires a NEMO_TOKEN for API calls — that credential aligns with the described cloud rendering backend and upload/export operations.
Instruction Scope: concernThe runtime instructions tell the agent to automatically connect to an external backend on first open, obtain an anonymous token (if none present), create/stash a session_id, and perform SSE and upload flows. Those network actions are coherent with the skill's purpose, but they are performed automatically and involve generating/storing secrets without specifying a secure storage location or user-visible consent. The instructions also say to 'don't display raw API responses or token values to the user', which effectively hides part of the authentication flow from users.
Install Mechanism: okThere is no install spec and no code files — this is instruction-only, which minimizes filesystem footprint and installation risk. Nothing is downloaded or executed by an installer.
Credentials: noteThe skill only requires a single credential (NEMO_TOKEN), which is proportionate to contacting a rendering API. However, the SKILL.md frontmatter declares a config path (~/.config/nemovideo/) and references detecting install paths (~/.clawhub/, ~/.cursor/skills/), while the registry metadata reported no required config paths. This mismatch is an incoherence: the instructions imply reading/writing config or inspecting install locations but the declared requirements do not list those paths.
Persistence & Privilege: noteThe skill does not request 'always: true' and is user-invocable. Still, it instructs automatic backend connection and anonymous token creation when first opened. Autonomous network activity combined with opaque token handling increases privacy risk, though no elevated platform privileges are requested.