Back to skill
Skillv1.0.0

ClawScan security

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

Scanner verdict

SuspiciousApr 22, 2026, 4:20 PM
Verdict
suspicious
Confidence
medium
Model
gpt-5-mini
Summary
The skill's instructions largely match a video-generation service and its single required credential (NEMO_TOKEN) is plausible, but there are minor metadata inconsistencies, an unknown upstream host, and privacy implications from uploading user video that you should understand before installing.
Guidance
This skill appears to do what it says (server-side video rendering) and only needs one API token, but pause before installing: 1) Source is unknown and there is no homepage — verify the vendor/reliability of mega-api-prod.nemovideo.ai before sending private footage. 2) Understand privacy/retention: uploaded videos will be sent to their servers; ask where files are stored, how long they are retained, and whether they are used for training. 3) Clarify the config-path discrepancy (~/.config/nemovideo/) with the publisher; ensure the skill won't read other local files. 4) Provide only the NEMO_TOKEN (or allow the skill to obtain an anonymous token) — do not give other unrelated credentials. 5) If you need stronger assurance, request an origin/homepage, documentation, or an official SDK/source repository to verify behavior. If you decide to proceed, monitor network activity and avoid uploading sensitive footage until you confirm retention and deletion policies.

Review Dimensions

Purpose & Capability
noteName/description claim: remote AI video generation from uploaded footage — the SKILL.md describes exactly that and requires a single API token (NEMO_TOKEN). This is proportionate. Minor inconsistency: top-level registry metadata listed no config paths, but the SKILL.md frontmatter declares a config path (~/.config/nemovideo/). That discrepancy should be clarified.
Instruction Scope
noteRuntime instructions direct the agent to obtain/use a token, create sessions, upload video files (up to 500MB), use SSE for editing operations, poll renders, and return download URLs from https://mega-api-prod.nemovideo.ai. These steps stay within the stated purpose. Important behavioral notes: user files are uploaded to a remote service (privacy/retention considerations); the SKILL.md also instructs to 'keep technical details out of the chat' which hides implementation details from users — benign for UX but worth flagging for transparency.
Install Mechanism
okInstruction-only skill with no install spec and no code files — lowest install risk. Nothing is downloaded or written by an installer as part of this skill package.
Credentials
noteOnly one credential (NEMO_TOKEN) is declared as required and that aligns with the described API usage. Two concerns: (1) SKILL.md frontmatter includes a config path (~/.config/nemovideo/) even though registry metadata lists none; (2) the skill instructs generating an anonymous token if NEMO_TOKEN is absent — that is a reasonable fallback but means the agent will call an auth endpoint and receive a short-lived token which the agent will then use to upload user videos. Neither is necessarily malicious but both should be understood before use.
Persistence & Privilege
okThe skill does not request always: true and does not claim to modify other skills or system-wide settings. Autonomous invocation (default) is enabled but not combined with unusually broad privileges.