Back to skill
Skillv1.0.0
ClawScan security
Text To Video In Runway · ClawHub's context-aware review of the artifact, metadata, and declared behavior.
Scanner verdict
SuspiciousApr 22, 2026, 8:57 AM
- Verdict
- suspicious
- Confidence
- medium
- Model
- gpt-5-mini
- Summary
- The skill's runtime instructions generally match a text-to-video service, but there are mismatches and behavioral choices (automatic anonymous-token creation, hidden technical details, and a metadata/configPaths inconsistency) that merit caution before installing or using it with sensitive data.
- Guidance
- This skill appears to be a wrapper for a nemovideo.ai text→video API and will upload media and call external endpoints. Before installing or using it: - Confirm you trust the domain (mega-api-prod.nemovideo.ai) and its privacy/storage policies, since uploads can be up to 500MB and may be retained on their servers. - Understand that if you don't provide NEMO_TOKEN the skill will automatically request an anonymous token and use it — the skill explicitly instructs the agent to hide those technical steps from the chat, so you may not see that exchange. - Ask the skill author to clarify the metadata/configPaths mismatch (~/.config/nemovideo/ present in SKILL.md but not in registry metadata) and whether any local config or tokens are written or read. - Check what metadata or headers (X-Skill-Source, X-Skill-Version, X-Skill-Platform) are sent with your uploads and whether that data is acceptable to share. - If you handle sensitive video/audio content, avoid uploading it until you have explicit confirmation about retention, access controls, and whether any obtained anonymous tokens persist beyond the session. If the answers are satisfactory (trusted service, clear retention/consent, and the metadata mismatch explained), the skill's behavior is reasonable for its purpose. If not, do not install or provide credentials.
Review Dimensions
- Purpose & Capability
- noteThe skill claims to perform text→video via a nemovideo.ai backend and the SKILL.md exclusively references that API and related endpoints — this is coherent. However, the registry metadata at the top of the package lists no required config paths while the SKILL.md frontmatter declares a configPaths entry (~/.config/nemovideo/), which is an internal inconsistency that should be clarified.
- Instruction Scope
- concernInstructions tell the agent to use NEMO_TOKEN if present or automatically obtain an anonymous token by POSTing to an external endpoint. The skill also instructs the agent to 'keep the technical details out of the chat' (i.e., hide network/auth steps from the user). These behaviors expand the agent's actions beyond simply calling an API with a user-provided key — they allow the skill to create credentials and perform network calls on the user's behalf without explicit, visible consent, and they require adding custom attribution headers to every request.
- Install Mechanism
- okNo install spec and no code files — the skill is instruction-only, so it does not write or execute bundled code on disk. This is the lowest-risk install mechanism.
- Credentials
- noteThe only declared required credential is NEMO_TOKEN (primaryEnv), which aligns with a cloud API integration. That said, the SKILL.md instructs the agent to create and use an anonymous token if NEMO_TOKEN is absent — effectively giving the skill the ability to obtain credentials automatically. Also note the frontmatter's configPaths entry (~/.config/nemovideo/) is not reflected in registry metadata, an inconsistency that could conceal expectations about local config access.
- Persistence & Privilege
- okThe skill does not request always:true and has no install-time persistence. It does describe session IDs and render job lifetimes on the backend, but nothing in the package requests permanent system-wide presence or modification of other skills.