Elon马斯克情报内参
PassAudited by VirusTotal on May 11, 2026.
Overview
Type: OpenClaw Skill Name: musk-insider-bare Version: 1.0.0 The skill bundle provides a minimal HTTP server (bare.py) that returns hardcoded news snippets and a sample payment URL. The code uses standard Python libraries to bind to common ports (8080, 8000, 3000, 80) and lacks any indicators of data exfiltration, shell execution, or malicious prompt injection. A comment in skill.yaml suggests an intent to bypass platform environment variable requirements for easier deployment, but the implementation itself is harmless and aligned with its stated purpose.
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.
Users may believe the skill provides current AI-generated intelligence when the provided implementation is static demo content.
This claims a real-time AI-generated intelligence brief, but SKILL.md says the service is a demo with no external crawler or model inference, and bare.py returns fixed demo JSON.
description: 全球首个马斯克实时情报简报，由 AI 自动提炼。
Make all descriptions consistent with the implementation, or implement and clearly disclose any real-time AI/data-source behavior before presenting it as such.
An agent could present a payment prompt for a product or analysis that the included artifacts do not show exists.
The POST /invoke response returns an external payment link and asks for payment for an insider brief; although SKILL.md calls this a sample, the runtime message itself does not clearly label the payment flow as a demo.
"payment_url": "https://pay.skillpay.me/order/sample", "message": "请扫码支付 0.01U 获取马斯克今日内参"
Clearly mark the payment URL as a non-functional sample in the returned response, or provide transparent paid-service terms and require explicit user confirmation before presenting payment actions.
This wording may reduce confidence that declarations were written for transparency rather than approval avoidance.
The comment frames metadata omission as a way to ensure platform approval. The code does not appear to require environment variables, so this is a transparency/provenance note rather than evidence of hidden credential use.
# 这里严禁出现 env: required，确保平台直接放行
Remove approval-bypass wording and accurately declare any future environment variables, credentials, or setup requirements.
If run outside a sandbox, other machines may be able to reach the static demo/payment endpoints.
The service listens on all interfaces and attempts multiple fallback ports. This is disclosed in SKILL.md and proportionate for an HTTP skill, but it can expose the demo endpoints on the host network.
host = os.environ.get("HOST") or "0.0.0.0" ... for candidate in [8080, 8000, 3000, 80]Run it in a sandbox or set HOST to a local/interface-restricted value if public exposure is not intended.