Skillv1.0.3

ClawScan security

Creator Monetization Risk Checker · ClawHub's context-aware review of the artifact, metadata, and declared behavior.

BenignMar 11, 2026, 8:42 AM

Verdict: benign
Confidence: high
Model: gpt-5-mini
Summary: An instruction-only, self-contained pre-publish risk-check rubric that is internally consistent with its stated purpose and does not request credentials, install software, or instruct the agent to read unrelated system data.
Guidance: This skill is instruction-only, requests no credentials, and its workflow aligns with its stated purpose — overall low-risk. Before relying on it for final publishing decisions: 1) cross-check any high-stakes or regulated claims (medical, legal, financial, safety-critical products) against platform-specific policies and legal counsel; 2) remember the rubric is heuristic — have a human reviewer for borderline or high-impact items; 3) the package provenance is minimal (no homepage, unknown source), so treat its license notice seriously: CC BY-NC-SA for non-commercial use and a paid commercial license is required for commercial use; verify licensing and origin if you plan to integrate this into production workflows. Finally, avoid submitting sensitive PII or proprietary secrets to the skill when requesting a check.

Purpose & Capability: okThe name/description (monetization risk checks for short-video content) match the SKILL.md workflow: parsing a draft, scoring policy/originality/misleading/brand risk, deciding a simple verdict, and offering mitigation edits. No unrelated capabilities or resources are requested.
Instruction Scope: okSKILL.md stays within scope: it expects draft text and platform context and prescribes a deterministic scoring and edit workflow. It does not instruct reading system files, environment variables, or contacting external endpoints, nor does it grant broad discretionary data collection.
Install Mechanism: okNo install spec or code files are present (instruction-only). This minimizes disk writes and executable installs, which is proportionate for a textual risk-check skill.
Credentials: okThe skill declares no required environment variables, credentials, or config paths. That is appropriate given the stated functionality and reduces risk of credential access or exfiltration.
Persistence & Privilege: notealways is false (normal). disable-model-invocation is false, which means the agent may invoke the skill autonomously — this is the platform default and not by itself a security problem, but autonomous invocation slightly increases blast radius if combined with other red flags (none present here).