Skillv1.0.0

ClawScan security

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

BenignApr 17, 2026, 4:14 PM

Verdict: Benign
Confidence: high
Model: gpt-5-mini
Summary: The skill's requirements and runtime instructions are consistent with a cloud-based video subtitle service: it asks for a single service token, uploads user video files to an external API, and contains no install or unrelated credential requests.
Guidance: This skill uploads your video files to an external service (mega-api-prod.nemovideo.ai) and uses a NEMO_TOKEN (or it can obtain a short-lived anonymous token) to operate. Before installing or using it: 1) Confirm you're comfortable uploading the specific videos (they will leave your machine). 2) Verify the NEMO_TOKEN you provide is the intended service token and not a different secret. 3) Be aware the skill may probe install paths (~/.clawhub, ~/.cursor/skills/) to populate an attribution header — this is a minor privacy action. 4) If you need to keep content private, do not upload it to third-party services. 5) If you want stronger guarantees, ask the skill author for a privacy/data-retention policy or a trustworthy homepage before using it.

Purpose & Capability: okThe skill is a video subtitle/encoding helper and only requests a single service credential (NEMO_TOKEN) and describes API calls for uploading videos, creating sessions, checking credits, and exporting rendered MP4s — all directly aligned with the described purpose.
Instruction Scope: noteInstructions explicitly direct the agent to upload user-provided video files (necessary for the feature), to call the external nemovideo.ai API, and to generate an anonymous token if no NEMO_TOKEN is present. The skill also instructs detecting the install path (~/.clawhub, ~/.cursor/skills/) to set an X-Skill-Platform header — this is a minor scope creep because it requires probing install paths (privacy-sensitive but not obviously malicious).
Install Mechanism: okThere is no install spec and no code files (instruction-only), so nothing is written to disk or downloaded during install — lowest-risk installation behavior.
Credentials: okOnly one environment variable is declared (NEMO_TOKEN) and it is the expected credential for the remote API. The SKILL.md does not request other unrelated secrets or system credentials.
Persistence & Privilege: okThe skill is not always-enabled and does not request elevated agent privileges. It can be invoked autonomously (platform default), which is expected for skills; no evidence it modifies other skills or system-wide configs.