Internet Failure Analysis Expert
Analysis
This is a coherent CVE and incident-report analysis skill with expected web-fetching behavior, and the provided artifacts do not show malicious or deceptive behavior.
Findings (3)
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.
Checks for instructions or behavior that redirect the agent, misuse tools, execute unexpected code, cascade across systems, exploit user trust, or continue outside the intended task.
url = sys.argv[1] ... response = requests.get(url, headers=headers, timeout=20)
The helper accepts a command-line URL and fetches it directly. This is expected for extracting public incident reports, but it is broad enough to fetch non-official or internal web addresses if the user or agent selects them.
requests>=2.31.0 beautifulsoup4>=4.12.0
The skill depends on external Python packages using minimum-version ranges. These are common, purpose-aligned dependencies, but non-pinned versions can resolve to newer package releases over time.
Checks for exposed credentials, poisoned memory or context, unclear communication boundaries, or sensitive data that could leave the user's control.
获取官方故障报告作为ground truth ... 对识别的官方报告URL，调用 `scripts/extract_webpage.py` 提取完整文本
The skill treats retrieved webpage/report text as authoritative context for the final analysis. That is central to the skill, but source errors, outdated reports, or prompt-like content in webpages could bias the generated report.