Skillv1.0.0

ClawScan security

Ai Image To Video Local · ClawHub's context-aware review of the artifact, metadata, and declared behavior.

Scanner verdict

SuspiciousApr 13, 2026, 10:29 AM

Verdict: suspicious
Confidence: high
Model: gpt-5-mini
Summary: The skill's description claims local/no-cloud processing, but its runtime instructions clearly send images and jobs to a cloud backend and perform actions (token creation, uploads, filesystem inspection) that don't match the 'local' promise.
Guidance: This skill claims to work 'locally' but actually uploads images and jobs to a cloud backend (mega-api-prod.nemovideo.ai). Before installing, decide whether you are comfortable with: (1) your images being uploaded to a third-party cloud service for processing; (2) the skill auto-generating an anonymous NEMO_TOKEN on your behalf (it will perform a network call to obtain a token if one is not present); and (3) the skill reading the skill file frontmatter and probing install paths to populate attribution headers (this may reveal local path information). If you expected offline/local-only processing, do not install/use this skill. If you accept cloud processing, verify the service's privacy/terms and only provide non-sensitive images. Ask the publisher for clarification about the 'local' claim and about what data is retained on the backend (jobs, images, logs) before proceeding.

Review Dimensions

Purpose & Capability: concernThe top-level description promises local processing 'without cloud uploads', but SKILL.md repeatedly describes a cloud render pipeline (upload endpoints, cloud GPU nodes, download URLs). That contradiction is a major incoherence: either the skill is mislabeled or it is misleading about where data goes. The frontmatter also lists a config path (~/.config/nemovideo/) while the registry metadata reported none, another small inconsistency.
Instruction Scope: concernRuntime instructions instruct the agent to read NEMO_TOKEN from the environment (expected) or automatically request an anonymous token via POST to https://mega-api-prod.nemovideo.ai/api/auth/anonymous-token (creates credentials), upload user images to cloud endpoints, poll state, and parse SSE. It also instructs reading the skill's YAML frontmatter and detecting install paths to set X-Skill-Platform — i.e., probing local install paths. Uploading user images and generating tokens are outside 'local' scope and may expose user data to an external service.
Install Mechanism: okInstruction-only skill with no install spec and no code files — lowest install risk. Nothing is written to disk by an installer. The operational risk comes from the runtime network calls described in SKILL.md rather than from an installer.
Credentials: concernThe skill requires a single credential (NEMO_TOKEN), which is appropriate for a cloud rendering API. However, it will auto-generate an anonymous token if none is present (network call that returns a token with limited credits) and it requires inclusion of attribution headers derived from local state (frontmatter and detected install path). Reading install paths and using local YAML frontmatter to populate headers can leak local environment details; auto-creating credentials without explicit user consent is also notable.
Persistence & Privilege: okThe skill is not always-enabled and does not request elevated persistent privileges. It stores and uses session_id during a session (expected). Autonomous invocation is allowed by platform default and not flagged on its own.