Skillv1.0.0

ClawScan security

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

Scanner verdict

SuspiciousApr 29, 2026, 7:08 PM

Verdict: suspicious
Confidence: medium
Model: gpt-5-mini
Summary: The skill's declared purpose (sharing video projects via a cloud editor) is plausible and its network/API calls line up, but there are a few inconsistencies and privacy implications you should understand before installing.
Guidance: This skill will upload the videos and media you hand it to an external service (mega-api-prod.nemovideo.ai) and will either use a provided NEMO_TOKEN or request an anonymous token on first use. Before installing, consider: 1) Do you trust nemovideo.ai with the videos you plan to upload? Don't use this for sensitive or private footage without confirming the service's privacy/retention policy. 2) If you prefer, provide your own NEMO_TOKEN rather than letting the skill acquire an anonymous token; ask the developer for documentation on token scope and lifetime. 3) Ask the publisher for source or a homepage and for clarification about the mentioned local config path (~/.config/nemovideo/) and whether the agent will read/write local files to set headers. 4) If you need stronger guarantees, test with non-sensitive sample files first and monitor network requests. The inconsistencies noted (missing homepage/source, frontmatter vs registry metadata mismatch, and automatic token creation) justify caution.
Findings: [NO_CODE_FILES] expected: The skill is instruction-only (SKILL.md) and the regex-based scanner found no code to analyze. This is expected, but it means the SKILL.md is the primary security surface to review.

Review Dimensions

Purpose & Capability: noteName/description match the instructions: the skill uploads video files to a cloud service and returns editor links. Requesting a NEMO_TOKEN is consistent with a cloud API. However, the SKILL.md frontmatter lists a config path (~/.config/nemovideo/) that the registry metadata did not declare, and the skill both requires NEMO_TOKEN and documents an anonymous-token acquisition flow — a mild inconsistency in how credentials are expected to be provided.
Instruction Scope: noteInstructions are prescriptive about creating sessions, uploading files, streaming SSE responses, polling render status, and including attribution headers — all coherent for a cloud video-processing tool. Important behavior: it will upload user-provided video/audio/media to https://mega-api-prod.nemovideo.ai and may automatically obtain an anonymous token if NEMO_TOKEN is not present. The instructions also infer install-paths (for X-Skill-Platform header), which implies the agent will inspect filesystem/paths or environment to set headers — the SKILL.md does not explicitly declare what path access is needed.
Install Mechanism: okInstruction-only skill with no install spec and no code files — nothing is written to disk by an installer. This is lower risk from an installation/execution standpoint.
Credentials: concernOnly NEMO_TOKEN is declared as the primary credential, which is proportionate. But the SKILL.md documents an automatic anonymous-token POST flow to obtain a temporary token when none is present; that behavior means the skill will make outbound auth requests on behalf of the user and provide a bearer token to the backend. The frontmatter also mentions a local config path (~/.config/nemovideo/), which was not declared elsewhere — this could mean the skill expects to read or write local config, which wasn't declared in the registry metadata.
Persistence & Privilege: okalways is false and the skill is user-invocable with normal autonomous invocation settings. The skill does not request permanent platform-wide privileges or claim to modify other skills' configuration.