Back to skill
Skillv1.0.0
ClawScan security
Text To Video Higgsfield Ai · ClawHub's context-aware review of the artifact, metadata, and declared behavior.
Scanner verdict
SuspiciousApr 17, 2026, 5:18 PM
- Verdict
- suspicious
- Confidence
- medium
- Model
- gpt-5-mini
- Summary
- The skill mostly matches a text→video cloud service (one API token required) but has a few inconsistencies and behaviors (automatic network setup, a hidden-config-path mention, and advice to suppress raw responses) that warrant caution before installing.
- Guidance
- This skill appears to be a straightforward text→video cloud integration, but exercise caution. Verify the backend domain (mega-api-prod.nemovideo.ai) and the service's reputation before uploading media or letting the skill create tokens. Ask the skill author to clarify why a local config path (~/.config/nemovideo/) is referenced (the registry metadata you received says none), how and where session tokens are stored, and whether the skill will make network calls automatically on first use. Prefer that the skill prompt you before creating or storing credentials, and avoid installing if you do not want uploads or automatic outbound connections to that third-party service. If you proceed, restrict and rotate any tokens and avoid sending sensitive files until you confirm the provider's privacy/retention policy.
Review Dimensions
- Purpose & Capability
- noteThe skill's name/description (text→video) align with the endpoints and NEMO_TOKEN usage. However there is an inconsistency between the registry requirements (reported earlier as 'no config paths') and the SKILL.md frontmatter which lists a config path (~/.config/nemovideo/). Asking for access to a local config directory is not obviously necessary for basic text→video operation and should be justified.
- Instruction Scope
- concernThe runtime instructions tell the agent to automatically connect to the backend the first time the skill is opened and to obtain an anonymous token if NEMO_TOKEN is not set. Automatic outbound network calls without explicit user confirmation and the guidance 'Don't display raw API responses or token values to the user' reduce transparency and increase risk of unnoticed data or credential flows. The instructions also require deriving an install-path-based header (X-Skill-Platform) which implies reading local paths — that access should be explicit and minimized.
- Install Mechanism
- okThis is an instruction-only skill with no install spec and no code files, so nothing is written to disk by an installer. That lowers installer-side risk.
- Credentials
- noteOnly a single credential (NEMO_TOKEN) is declared, which is appropriate for a cloud video service. But SKILL.md contains a frontmatter configPaths entry (~/.config/nemovideo/) suggesting the skill may read local config for tokens or state; the public registry metadata given to you earlier said no config paths. This mismatch should be resolved: reading a user config directory is more access than just needing one API token.
- Persistence & Privilege
- okalways is false and the skill is user-invocable; it does not request permanent/always-on privileges. It does instruct storing a session_id for subsequent calls, which is normal for session-based APIs.