Skillv1.0.1

ClawScan security

X Mobile Longshot / X 真机感长截图导出 · ClawHub's context-aware review of the artifact, metadata, and declared behavior.

Scanner verdict

SuspiciousMar 14, 2026, 5:28 AM

Verdict: suspicious
Confidence: medium
Model: gpt-5-mini
Summary: The skill's code matches its stated purpose (rendering X pages to mobile-style screenshots), but the package metadata omits important runtime requirements and the script executes local binaries (python3) — these mismatches deserve attention before use.
Guidance: This skill appears to do what it says (capture mobile-style screenshots of X pages), but the package metadata omits key runtime requirements. Before installing or running it: 1) Ensure the runtime has Node.js and the Playwright Node package (and its browsers) installed, and has python3 with Pillow available — the script calls python3 via spawnSync to process images. 2) Run it in an isolated environment (no sensitive credentials mounted) because it will fetch arbitrary URLs you supply and write files to disk. 3) Review the bundled script if you need stricter safety: it performs DOM clicks/removals (to dismiss overlays) and executes local Python code — both expected for this task but worth auditing. 4) Prefer adding explicit dependency/install steps (Playwright browser install, pip install Pillow) to the skill manifest or run within a container where you control installed packages. If you need higher assurance, ask the author to declare required binaries and provide an install spec rather than relying on implicit environment setup.

Review Dimensions

Purpose & Capability: noteThe skill's name/description align with the included script: it uses Playwright to render X (Twitter) pages, cleans overlay nodes, captures screenshots, and converts PNG→PDF. However the registry metadata lists no required binaries or environment requirements even though the runtime clearly needs Node.js (to run the script), the Playwright library (and its browser downloads), Python3, and Pillow. That gap is an inconsistency (likely sloppy packaging) that users should be aware of.
Instruction Scope: okSKILL.md and the script stay within the stated purpose: they load a user-provided URL, manipulate the page DOM (click/ remove overlays), capture screenshots, and post-process images locally. The script does page DOM clicks/removals and runs local Python code via spawnSync, but it does not call external endpoints or read unrelated system files or environment variables.
Install Mechanism: concernThere is no install spec (instruction-only + included script). That is lower risk than arbitrary remote downloads, but the absence of declared installs/dependencies is problematic because Playwright typically requires installing browser binaries and Python/Pillow must be available. The script will call python3 and expects PIL; the skill doesn't document how to provision those, so operators may be surprised by missing/automatic installs or by heavy Playwright browser downloads.
Credentials: okThe skill requests no environment variables or credentials, and the code does not attempt to access secrets or unrelated config paths. It does spawn local processes (python3) and writes output files under user-provided paths, which is consistent with its purpose.
Persistence & Privilege: okThe skill is not always-enabled and does not request elevated/persistent agent privileges. It does not modify other skills or system-wide agent settings. Autonomous invocation remains possible (platform default) but is not combined with other concerning flags here.