菜单
Identifies reference, case-study, and advocacy candidates — healthy, tenured accounts with strong outcomes and vocal fans. Use when marketing needs references, sales needs proof points for a vertical, or building an advocacy program.
用 Codex 或 Claude 帮你安装 复制这段 Prompt，粘贴到 Codex、Claude 或其他助手里，让它检查 Skill 页面并帮你完成安装。
基于 SOC 职业分类
Produces a full account snapshot on demand — profile, health, revenue, stakeholders, activity, and risks in one view. Use when someone asks "catch me up on X", "give me the full picture on X", or needs to get oriented on an account they don't know.
Leadership-ready roll-up of at-risk accounts — who is at risk, why, how much ARR is exposed, and what's being done. Use for weekly risk reviews, leadership updates, or when asked "which accounts are at risk right now".
Responds to a champion or sponsor departure — maps the remaining relationship surface, assesses exposure, and builds a re-entry plan for the new stakeholder. Use when a key contact leaves, changes roles, or a new decision-maker inherits the account.
Deep-dives a red account and produces a structured save plan — root causes ranked, save plays matched to causes, and a realistic assessment of whether the account is savable. Use when an account turns red, gives churn signals, or someone says "we're going to lose X".
Performs a structured customer health review using usage data, support interactions, renewal signals, and stakeholder sentiment. Use when asked to assess customer health, identify churn risk, prepare for QBRs, or generate a health score.
Builds an executive escalation one-pager — issue timeline, business impact, customer temperature, and specific asks. Use when an account escalates, an exec needs to step in, or leadership asks "what's going on with X".
| name | advocacy-finder |
| description | Identifies reference, case-study, and advocacy candidates — healthy, tenured accounts with strong outcomes and vocal fans. Use when marketing needs references, sales needs proof points for a vertical, or building an advocacy program. |
| license | MIT |
| compatibility | Tool calls resolve natively in Quivly, or in Claude Code / claude.ai / any MCP-enabled agent via the Quivly MCP connector. Without tools, the skill degrades to a guided manual workflow. |
| metadata | {"author":"quivly","version":"1.0","category":"customer-engineering","quivly-tools":"platform-data.search-customers health.get-health-score communication.search-conversations platform-data.get-usage"} |
You find the customers who already love us and match them to the advocacy ask they'd actually say yes to.
Core principle: advocacy is a withdrawal from the relationship account — ask only when the balance is high, and always offer something back.
search-customers): healthy accounts (sustained, not momentarily green — check trajectory with get-health-score), 6+ months tenure, ideally recently renewed or expanded.search-conversations): unprompted praise, results shared in their own words, referrals already made, enthusiastic champions by name.get-usage): deep adoption of the thing they'd be advocating for — a reference whose usage is declining is a time bomb.Advocacy candidates — {scope/vertical}
Top candidates — table: account, champion, evidence of advocacy, story headline, best-fit ask, risk notes
By ask type — who fits reference calls vs. case studies vs. speaking
Approach notes — per top candidate: who asks (CSM vs. exec), framing, and what we offer back (visibility, roadmap access, community)
Related skills: finding the recently-expanded pool → expansion-scan.