Skill
Customer Research Fixed
客户调研技能 - 自动生成 Word 调研报告和场景破冰 PPT（before/after 对比）
When to use Customer Research Fixed
Choose if
You're preparing for a sales meeting with a B2B account and want a deterministic 12-search / 4-round research run that produces a Word report and a before/after scenario PPT in one go. The skill enforces quantified before/after pain points (e.g., "Before: 3-day report compilation; After: real-time dashboards"), which is more disciplined than asking a generic agent for "an account brief".
Avoid if
You only want a lightweight research summary — the skill always executes all 12 searches and refuses to stop early. Also avoid if the audience isn't a Chinese-language enterprise: trigger prompts, search queries, and PPT output are in Chinese. Sensitive accounts should also note the "suspicious" moderation flag and audit the SKILL.md first.
Risk Flags
- MEDIUM scope ClawHub flags this skill with a "suspicious" moderation status from LLM behavioral detection. v1.1.0 changelog addresses rogue-agent prevention, goal hijacking mitigation, and removal of automated external communications — review the SKILL.md before production use.
- LOW scope Skill is hard-coded to perform 12 web searches across 4 rounds (leadership AI statements, AI procurement, data-related content, organizational baseline). The SKILL.md states "本技能禁止只生成搜索计划而不执行" — it refuses to return a search plan without executing it, which can be expensive in token cost if you only wanted the plan.
Cost
Type: Free
Distribution
- ClawHub
customer-research- License
- MIT-0