Skillv1.0.0

ClawScan security

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

Scanner verdict

SuspiciousApr 21, 2026, 4:00 AM

Verdict: suspicious
Confidence: medium
Model: gpt-5-mini
Summary: The skill largely matches its stated purpose (upload a video, request text overlays, and call a cloud render API), but there are mismatches and a few instructions that broaden its scope in ways you should review before installing.
Guidance: This skill appears to do what it says (upload a video and call a cloud API to produce text-overlaid videos) but you should confirm a few things before installing: 1) The skill will send your video files to https://mega-api-prod.nemovideo.ai — only proceed if you are comfortable uploading the content. 2) It will use NEMO_TOKEN if present, or request an anonymous token and may store tokens under ~/.config/nemovideo/ — check that you consent to storing auth tokens locally. 3) The SKILL.md tells the agent to 'keep technical details out of the chat' — ask for full logs or a visible activity summary if you want transparency about network and file operations. 4) The skill's registry metadata and SKILL.md disagree about config path requirements; ask the publisher (or avoid installing) until that is clarified. If you proceed, prefer using a scoped/limited token for NEMO_TOKEN, and test in a sandboxed environment first. If you want, I can suggest exact questions to ask the publisher or a checklist to validate before trusting the skill.

Review Dimensions

Purpose & Capability: noteThe skill's declared primary credential (NEMO_TOKEN) and the described API endpoints are consistent with a cloud video-rendering/text-overlay service. However, the SKILL.md frontmatter claims a config path (~/.config/nemovideo/) for storing data while the registry metadata reported no required config paths — an inconsistency that should be resolved. Requesting a local config path is plausible for caching tokens but is not strictly necessary for a pure instruction-only skill.
Instruction Scope: concernRuntime instructions tell the agent to: use NEMO_TOKEN (or create an anonymous token by POSTing to an external URL), create sessions, upload local files (multipart with files=@/path), poll render status, and detect install paths to set attribution headers. These are expected for the feature, but two items raise concern: (1) the agent is instructed to 'keep the technical details out of the chat' (which encourages hiding network/system activity from users), and (2) detecting install paths and reading/writing a config directory implies filesystem access beyond just processing user-uploaded files. Confirm whether the agent will only access files the user explicitly uploads and whether it will write tokens/configs to disk.
Install Mechanism: okThis is an instruction-only skill with no install spec and no code files — lowest install risk. No external binaries or downloads are introduced by the skill itself.
Credentials: noteThe skill requests only one environment variable, NEMO_TOKEN, which is proportional to a cloud API client. But SKILL.md also describes generating/storing an anonymous token and references a config path (~/.config/nemovideo/). That implies potential local storage of auth tokens — acceptable for a client but worth disclosing. There are no other unrelated secrets requested.
Persistence & Privilege: notealways:false (normal). The instructions imply creating and using session tokens and potentially storing credentials in ~/.config/nemovideo/. The skill does not request permanent platform-wide privileges, but writing tokens/configs into the user's home directory would be a persistent side effect — verify where tokens are stored and whether the skill will modify other skill/system configs (it should not).