Nsfw Detection

PassAudited by VirusTotal on May 12, 2026.

Overview

Type: OpenClaw Skill Name: nsfw-detection Version: 1.0.0 The skill bundle describes legitimate NSFW detection and URL reputation functionalities. However, the `SKILL.md` file lists two different npm packages for installation: `@clawhub/nsfw-detector` and `@raghulpasupathi/nsfw-detection`. This discrepancy introduces a potential supply chain vulnerability, as a user might inadvertently install a package from a different scope (`@raghulpasupathi`) which could be a typosquatting target or a compromised package, even if the document itself does not exhibit malicious intent.

Findings (0)

Artifact-based informational review of SKILL.md, metadata, install specs, static scan signals, and capability signals. ClawScan does not execute the skill or run runtime probes.

What this means

Installing the referenced package would trust code that was not reviewed here, although this is a normal setup pattern for an npm-based skill.

Why it was flagged

The skill is instruction-only but directs users to install external npm packages whose code is not included in the provided artifacts.

Skill content
npm install @raghulpasupathi/nsfw-detection
Recommendation

Verify the npm package publisher, source, version, and install scripts before installing or using it.

What this means

Users could over-rely on the filter for sensitive safety decisions if they treat the stated accuracy as guaranteed.

Why it was flagged

The skill makes safety-relevant accuracy claims and child-protection use-case claims, but the provided artifacts do not include test data, model files, or code to substantiate them.

Skill content
Accuracy: 92-95% on explicit content, 75-80% on suggestive
Recommendation

Treat the accuracy numbers as unverified documentation, test the model in your own environment, and keep human oversight for child-safety or moderation use cases.