From agentic-ai-skills
Assists pre-sales engineers with customer profiling, Ghost Deck generation, proposal creation, expert self-reviews, talking points, objection handling, opportunity scoring, and follow-up plans targeting government, top enterprises, and tech firms.
npx claudepluginhub agenticaiplan/agenticaiskills --plugin agentic-ai-skillsThis skill uses the workspace's default tool permissions.
面向三类客户(政府、行业头部企业、技术型企业)的全链路售前工作流 Skill。
config/client_profiles.yamlconfig/gate_rules.yamlconfig/objection_database.yamlconfig/pain_point_map.yamlevals/evals.jsonreferences/enterprise_sop.mdreferences/government_sop.mdreferences/objection_playbooks.mdreferences/tech_sop.mdscripts/collect_client_info.pyscripts/generate_followup.pyscripts/generate_ghost_deck.pyscripts/generate_proposal.pyscripts/generate_talking_points.pyscripts/handle_objections.pyscripts/score_opportunity.pyscripts/self_review.pyutils/gate_check.pyutils/input_parser.pyutils/output_formatter.pyGenerates sales playbooks: outbound sequences, discovery call guides, objection handling scripts, demo frameworks, and proposal templates. Captures ICP context for tailored outputs.
Guides sales activities: crafting outreach emails, preparing demos, handling objections, managing pipelines, and building CRM workflows for customer acquisition.
Qualify sales deals using MEDDIC, BANT, SPICED, or CHAMP frameworks with structured scoring and gap analysis. Use this skill whenever a rep needs to qualify a deal, assess deal health, decide whether to invest more time in an opportunity, says "should I pursue this deal", "qualify this opportunity", "MEDDIC score", "BANT check", or asks about deal qualification criteria. Also trigger when a manager reviews pipeline quality or when someone mentions deal scoring, opportunity assessment, or go/no-go decisions.
Share bugs, ideas, or general feedback.
面向三类客户(政府、行业头部企业、技术型企业)的全链路售前工作流 Skill。
| 命令 | 功能 | 适用时机 |
|---|---|---|
/profile | 对话式客户信息采集,生成画像 | 拜访前 |
/ghost | Ghost Deck(目录+Action Title) | 方案设计阶段 |
/proposal | 逐章生成方案内容 | Ghost Deck确认后 |
/review | 专家视角自审(GO/NO-GO) | 方案提交前 |
/talking-points | 演讲要点生成 | 拜访前1天 |
/objections | 异议处理准备 | 拜访前 |
/score | 商机 Close Readiness 评分 | 任何阶段 |
/followup | 跟进行动计划 + 会议纪要草稿 | 会后1小时内 |
/presales-flow | 完整流程(串联全部模块) | 全新商机 |
/presales-flow --mode quick # 快速报价(15分钟,1页摘要)
/presales-flow --mode standard # 标准方案(2小时,5-10页)
/presales-flow --mode tender # 投标响应(分阶段,20-50页)
[!CAUTION]
- 不编造资质证书、业绩案例、客户名称 → 用**【待补充】**占位
- 不虚构技术参数、Benchmark数据 → 用**【待核实】**占位
- 不编造任何价格/成本/合同数字 → 用**【待填】**占位
- 不直接贬低任何竞争对手
- 方案质量门禁由
utils/gate_check.py程序判断,不口头宣布"通过"
/profile执行脚本:
python3 scripts/collect_client_info.py [初始客户信息]
工作方式(借鉴 csm-ebr-qbr-prep-skill 逐一问答模式):
根据客户类型加载对应参考文件:
references/government_sop.md 的摸底维度references/enterprise_sop.mdreferences/tech_sop.md/ghost执行脚本:
python3 scripts/generate_ghost_deck.py [客户类型]
Ghost Structure First 原则(借鉴 mbb-decks):
以上是方案的叙事目录,整体逻辑是否流畅?
有哪些章节需要调整顺序、删除或增加?
确认后我将逐章展开内容。
Action Title 规范(每页标题必须符合):
/proposal执行脚本:
python3 scripts/generate_proposal.py [客户类型] [模式]
逐章生成流程:
禁止区强制执行:
/review执行脚本:
python3 scripts/self_review.py [客户类型]
虚拟专家团队(借鉴 expert-review-panel 反谄媚机制):
反谄媚规则(自动执行):
输出:
在提交/发送方案前执行:
python3 utils/gate_check.py outputs/{客户名}_{日期}/proposal.md
检查结果写入 gate_result.json。以程序输出为准,我不口头宣布通过。
检查项:占位符残留 / 可疑数字 / 竞品不当评价 / 敏感信息泄露
/talking-points执行脚本:
python3 scripts/generate_talking_points.py [客户类型]
输入:确认的方案目录 + 参会人员角色 + 会议时长 + 会议目标
输出:
/objections执行脚本:
python3 scripts/handle_objections.py [客户类型] [已知顾虑]
从 config/objection_database.yaml 匹配最相关的 TOP5 异议,输出:
/score执行脚本:
python3 scripts/score_opportunity.py [客户类型] [情况描述]
8项 Close Readiness 指标评分:
/followup(会后执行)执行脚本:
python3 scripts/generate_followup.py [会议要点] [客户反应] [决策周期]
输出:
| 文件 | 加载时机 |
|---|---|
references/government_sop.md | 客户类型为政府时 |
references/enterprise_sop.md | 客户类型为头部企业时 |
references/tech_sop.md | 客户类型为技术型企业时 |
references/objection_playbooks.md | 执行 /objections 时 |