Skillv1.0.0

ClawScan security

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

BenignApr 20, 2026, 2:24 PM

Verdict: benign
Confidence: high
Model: gpt-5-mini
Summary: The skill's requirements and runtime instructions are coherent with an AI video-rendering service: it needs a NEMO_TOKEN (or will obtain an anonymous one), uploads user media to a remote render backend, and does not request unrelated credentials or install arbitrary software locally.
Guidance: This skill will upload any images, video, or audio you provide to a third‑party backend (mega-api-prod.nemovideo.ai) for processing and requires a NEMO_TOKEN (or will request an anonymous token on your behalf). Before installing/using: (1) Confirm you are comfortable with your media and any transcript being sent to that external service; avoid uploading sensitive or confidential content. (2) Verify the backend domain is legitimate for the vendor you expect. (3) Be aware the skill reads its own SKILL.md frontmatter and inspects common install paths to set attribution headers — this can reveal which client/platform you use. (4) The registry metadata and SKILL.md disagree about config path requirements (minor); if you care where local state is stored, check ~/.config/nemovideo/ after first run. Overall the permissions are proportional to its purpose, but review the service's privacy/terms and avoid sending sensitive data.

Purpose & Capability: noteName/description (animated video generation) matches the declared primary credential (NEMO_TOKEN) and the runtime API endpoints. One minor metadata mismatch: the registry-level summary listed no required config paths, while the SKILL.md frontmatter includes a config path (~/.config/nemovideo/) — plausible but inconsistent and worth confirming.
Instruction Scope: noteSKILL.md focuses on the render workflow (session creation, SSE chat, upload, export and polling) which is within scope. It instructs the agent to read the skill's own frontmatter and inspect common install paths (e.g., ~/.clawhub/, ~/.cursor/skills/) to set an X-Skill-Platform header — functionally reasonable but a privacy/detail leak risk because it reads local path information and sends that platform attribution to the backend.
Install Mechanism: okInstruction-only skill with no install spec or code files; nothing is written to disk by an installer. Lowest-risk install mechanism.
Credentials: okOnly requests a single service credential (NEMO_TOKEN) and documents how to obtain an anonymous token if missing. No unrelated secrets or multiple credentials are requested. It does declare a config path for its own use, which is proportionate to a client that may cache session state.
Persistence & Privilege: okalways:false and normal autonomous invocation are set. The skill does not request permanent system-wide privileges or attempt to modify other skills' configs.