Content Researcher
Analysis
The skill is coherent and purpose-aligned, using web search and optional summarization to create reports, with ordinary cautions around external dependencies, LLM summaries, and the chosen output file.
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.
subprocess.run(["claw", "tools", "web_search", "--query", query, ...]); ... with open(output_path, 'w', encoding='utf-8') as f: f.write(report)
The skill invokes local CLI tools and writes a report to a user-chosen path. This is central to its purpose and disclosed, but users should be aware it performs tool actions and can overwrite the selected output file.
需要安装这两个工具（均通过 clawhub 安装） ... clawhub install summarize
The skill relies on separately installed claw/summarize commands. This is disclosed and purpose-aligned, but the trustworthiness of those external tools affects the skill.
Checks for exposed credentials, poisoned memory or context, unclear communication boundaries, or sensitive data that could leave the user's control.
包含：所有搜索结果的标题、来源、链接、摘要 ... 如果启用 `--summarize`，还有 AI 生成的总结
The skill saves retrieved web snippets and AI-generated summaries into a reusable report. Online source content can be inaccurate, biased, or adversarial, even though collecting it is the intended function.