读取调研outline,为每个item启动独立agent进行深度调研。禁用task output。
74%
Does it follow best practices?
Run evals on this skill
Adds up to 20 points to the overall score
Low
Low-risk findings worth noting
Fix and improve this skill with Tessl
tessl review fix ./skills/research-zh/research-deep/SKILL.mdThe skill exposes the agent to untrusted, user-generated content from public third-party sources, creating a risk of indirect prompt injection. This includes browsing arbitrary URLs, reading social media posts or forum comments, and analyzing content from unknown websites.
SKILL.md的runtime流程在Step 3启动了web-search-agent并用“调研 {item_related_info}”来执行,表明LLM会进行基于外部搜索的自由文本阅读/摄取,因此存在被外部(搜索结果/页面)注入的间接提示风险。
Weizhena/Deep-Research-skills
6ce38f6
If you maintain this skill, you can claim it as your own. Once claimed, you can manage eval scenarios, bundle related skills, attach documentation or rules, and ensure cross-agent compatibility.