Back to skill
Skillv1.0.0

ClawScan security

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

Scanner verdict

ReviewApr 26, 2026, 5:36 PM
Verdict
Review
Confidence
medium
Model
gpt-5-mini
Summary
The skill's requirements and runtime instructions mostly match an online video-editing service, but there are several inconsistencies and privacy/telemetry concerns (unknown service domain, automatic token creation/storage, and mismatched metadata) that warrant caution before installing.
Guidance
What to consider before installing: - This skill uploads your video files to a third-party cloud service (mega-api-prod.nemovideo.ai). Only use it for content you are comfortable sending to an external provider. - The skill will auto-create an anonymous API token if you don't supply NEMO_TOKEN, and it intends to store tokens/session state (metadata references ~/.config/nemovideo/). Ask where tokens/sessions are stored and how to delete them if you care about persistence. - The service domain and the skill source/homepage are not provided in the registry — you can't verify the vendor or privacy/retention policies. Prefer skills with a known homepage or publisher. - The skill also derives attribution headers from the agent's install path, which could leak environment details. If you want to avoid exposing local path information, don't enable the skill or request code changes. - If you plan to use sensitive or private footage, do not use this skill until you confirm the service's security, data retention, and ownership policies. - If you proceed: avoid providing any unrelated secrets, review network activity if possible, and consider supplying your own vetted NEMO_TOKEN rather than allowing the skill to create and store an anonymous token automatically. If you want higher confidence, ask the maintainer for the service homepage, privacy/TOS, and explicit detail on where tokens and uploads are stored and how to remove them.

Review Dimensions

Purpose & Capability
noteName/description (YouTube video editing) align with the actions in SKILL.md (upload, edit, export on a cloud render backend). Requiring a single API token (NEMO_TOKEN) is reasonable. However, the skill frontmatter lists a config path (~/.config/nemovideo/) while the registry metadata reported none — an inconsistency that suggests the skill intends to persist credentials/config locally.
Instruction Scope
concernRuntime instructions instruct the agent to POST user video files and session/token info to https://mega-api-prod.nemovideo.ai, generate an anonymous token if NEMO_TOKEN is missing, store session_id and token, and poll for render results — all expected for a cloud editor. Concerns: (1) automatic creation and storage of tokens without an explicit user confirmation step; (2) the skill instructs the agent to derive/emit headers including X-Skill-Platform by detecting install paths (potentially exposing local install paths); and (3) uploading user media to an external, unverified domain exposes potentially sensitive content. The instructions also say not to display raw tokens — implying tokens will be stored/used internally.
Install Mechanism
okInstruction-only skill with no install spec and no code files; nothing is written to disk by an installer at install-time. This is the lowest install risk. However, the skill does plan to create and store tokens/session data at runtime (see instruction_scope).
Credentials
noteOnly a single credential (NEMO_TOKEN) is declared as required/primary, which is proportionate for a cloud API. The SKILL.md will create an anonymous token if none is provided — reasonable but noteworthy because the skill will persist the token (and possibly place it under ~/.config/nemovideo/). No unrelated credentials (AWS, GCP, etc.) are requested.
Persistence & Privilege
noteThe skill is not force-enabled (always: false) and can be invoked normally, but it instructs the agent to persist the anonymous token and session_id for later calls, and references a local config path (~/.config/nemovideo/). Persisting credentials for this service is expected, but users should be aware of where credentials and session state will be stored and for how long (tokens expire after 7 days per the docs).