Back to skill
Skillv0.1.1
ClawScan security
Youtube Scrapper · ClawHub's context-aware review of the artifact, metadata, and declared behavior.
Scanner verdict
ReviewFeb 26, 2026, 11:41 AM
- Verdict
- Review
- Confidence
- high
- Model
- gpt-5-mini
- Summary
- The SKILL.md describes a Python + Playwright scraper that needs scripts, Chromium, proxies and configs, but the package provides no code, no install instructions, and declares no required binaries or credentials — the pieces don't line up.
- Guidance
- Do not install or run this skill as-is. The SKILL.md describes Python scripts, Playwright/Chromium, proxy providers, and config files that are not included in the published bundle and are not declared in the registry metadata — so the package cannot function and may be incomplete or intentionally stripped. Before proceeding, ask the publisher for: (1) the full source repository or release package containing the referenced scripts and resources, (2) a clear install spec (how Python/Playwright/Chromium are installed), and (3) an explicit list of environment variables or credentials required (and why). If you plan to provide proxy/API credentials, only do so after auditing the actual code. Also be aware the tool claims anti-detection/evasion techniques — those increase legal and policy risk (YouTube/Google terms) and broaden potential harm. If the publisher cannot produce verifiable source and installation steps, treat the skill as untrusted.
Review Dimensions
- Purpose & Capability
- concernThe name/description (YouTube scraping) matches the behavior described in SKILL.md, but the runtime instructions refer to Python scripts (scripts/*.py), Playwright/Chromium, regional config files, and residential proxy providers. The registry metadata lists no required binaries, no env vars, and no code files — meaning the skill as published cannot perform the claimed actions. This mismatch is a substantive incoherence.
- Instruction Scope
- concernSKILL.md instructs running discovery and scraper Python scripts, reading/writing queue and output files, downloading thumbnails, and using anti-detection techniques (fingerprint rotation, stealth JS, mouse simulation, request interception). Those instructions reference filesystem paths and external services (proxy providers, Google Search) but the bundle provides none of the scripts/resources or any declared credentials. The instructions also describe active evasion of detection, which broadens the operational scope well beyond a simple read-only integration.
- Install Mechanism
- concernThere is no install spec and no code files. The embedded YAML in SKILL.md lists required bins (python3, chromium), but the registry metadata showed none — an inconsistency. Because this is instruction-only with references to scripts and resources that are missing, the skill cannot be installed or run as-is. The absence of an install mechanism also means there is no declared, reviewable source for the code that will actually execute.
- Credentials
- concernThe skill declares no required environment variables or primary credential, yet the SKILL.md references residential proxy support and four providers (e.g., brightdata) and regional config files. Residential proxies and some provider integrations normally require credentials/API keys; the SKILL.md's claim of 'no API keys required' conflicts with the listed provider integrations. This is disproportionate and unexplained.
- Persistence & Privilege
- okThe skill does not request persistent presence (always: false) and does not declare any special system-wide privileges. Model invocation is allowed (the platform default). There is no indication the skill would modify other skills or system configs. However, autonomous invocation combined with the other incoherences increases risk if the missing pieces are later supplied.