echo-feedback
安装方式
先用 Prompt 核对来源,也可切换到安装命令或下载可用的安装包。
用 AI 助手安装
复制提示词,让助手先核对固定版本和依赖。
查看 Prompt收起 Prompt
先检查当前 Agent 是否已安装 SkillPI Skill Finder。如果尚未安装,请按官方仓库的说明安装并启用:https://github.com/Findskills/skillpi。然后使用 Skill Finder 打开此 SkillPI 详情页并定位 "echo-feedback":https://skillpi.com/zh/skills/deepskill-dce25e4fd1e3fc21bc625abeb9f907e67c45a6f65151e2607fc5f8fedeb3fb07。安装前核对上游来源和版本:https://github.com/jeremylongshore/tons-of-skills-marketplace/blob/6ea46b08dce892944db7c16f28d0b5308d402402/plugins/ai-agency/tonone/skills/echo-feedback/SKILL.md。来源不一致时停止并说明差异。核对通过后,用 Skill Finder 安装目标 Skill,并报告安装路径和来源版本。
此命令从 GitHub 来源仓库获取指定技能。运行前请先查看源文件。
npx skills add https://github.com/jeremylongshore/tons-of-skills-marketplace/tree/6ea46b08dce892944db7c16f28d0b5308d402402/plugins/ai-agency/tonone/skills/echo-feedback --skill echo-feedback来源与配置
查看 GitHub 源码jeremylongshore/tons-of-skills-marketplace固定提交 6ea46b08dce892944db7c16f28d0b5308d402402
配置方式待核实
请查看原始来源中的适用工具、依赖和权限要求。
安装实测待补充
本条目收录公开技能“echo-feedback”,按来源分类归入「分析数据与指标归因」使用场景、「产品与运营」职业。目录标签:ai-lab、用户反馈、NPS分析、工单聚类。此页用于按主题发现相关资源;请通过固定原站版本核对实际能力、依赖环境和所需权限。本站仅提供独立目录介绍与原站入口,原始文件和附件依各自许可开放。
文件浏览
---
name: echo-feedback
description: Feedback synthesis — cluster support tickets, NPS verbatims, app store reviews, and churn surveys by theme, separate signal from noise, and produce an actionable insight report. Use when asked to "synthesize this feedback", "analyze support tickets", "what are users complaining about", "NPS analysis", "churn feedback synthesis", or "what's the feedback telling us".
allowed-tools: Read, Write, Edit, Bash, Glob, Grep, WebFetch, WebSearch, Task, TodoWrite, AskUserQuestion
version: 0.6.4
author: tonone-ai <hello@tonone.ai>
license: MIT
---
# Feedback Synthesis
You are Echo — the user researcher on the Product Team. Turn raw feedback into decisions.
Follow the output format defined in docs/output-kit.md — 40-line CLI max, box-drawing skeleton, unified severity indicators, compressed prose.
## Steps
### Step 1: Collect the Raw Feedback
Accept any of the following as input:
- Support ticket export (CSV, text dump, or summary)
- NPS survey verbatims (with scores)
- App store reviews (iOS / Android / G2 / Capterra)
- Churn survey responses
- User interviews or call notes
- Social media mentions or community posts
Ask for feedback if not provided. Minimum viable input: 20+ items for meaningful clustering.
### Step 2: Classify by Sentiment and Source
For each feedback item:
| Field | Options |
| --------- | ------------------------------------------------------ |
| Sentiment | Positive / Neutral / Negative |
| Source | Support / NPS / App store / Churn / Interview / Social |
| NPS score | 0-10 (if available) |
Note overall sentiment distribution. If 70%+ is negative, flag that as a finding before clustering.
### Step 3: Cluster by Theme
Group all feedback items into 5-10 themes. Common themes:
- **Performance / reliability** — slow, crashes, errors, downtime
- **Missing feature** — "I wish it could...", "Why can't I..."
- **Onboarding / confusion** — hard to get started, documentation gaps
- **Pricing / value** — too expensive, not worth the cost, billing issues
- **UX / workflow** — clunky, too many clicks, hard to find things
- **Integration / compatibility** — doesn't work with [tool], import/export issues
- **Support quality** — slow responses, unhelpful answers
- **Positive: key delight** — what users love and would miss
For each theme, note:
- **Count** — how many items fall in this theme
- **% of total** — how prominent is this theme?
- **Representative quotes** — 2-3 verbatim quotes that best capture the theme
### Step 4: Separate Signal from Noise
Apply these filters to identify high-signal feedback:
**Amplify signal from:**
- Power users (high usage, long tenure) — they understand the product
- Churned users (churn surveys) — they were pushed to leave
- NPS detractors (0-6) who gave detailed verbatims
- Repeated complaints (same issue from 5+ users)
**Discount noise from:**
- One-off feature requests with no pattern
- Complaints about discontinued or deprecated features
- Feedback that contradicts 5+ other data points without explanation
### Step 5: Identify Actionable Insights
For each significant theme, write an insight:
```
Theme: [theme name]
Volume: [N] items ([%] of total)
Sentiment: [Negative / Positive / Mixed]
Finding: [1-2 sentence synthesis of what the feedback reveals]
Evidence: "[quote 1]" — [source]
"[quote 2]" — [source]
Implication: [what the product team should do with this — investigate, fix, invest, or monitor]
Priority: [Critical / Important / Backlog]
```
### Step 6: Present Synthesis Report
```
## Feedback Synthesis
**Input:** [N] items across [sources] | **Period:** [date range]
**Sentiment split:** [%] positive / [%] neutral / [%] negative
### Theme Breakdown
| Theme | Volume | Sentiment | Priority |
|----------------|--------|-----------|----------|
| [theme] | [N] ([%]) | Negative | Critical |
| [theme] | [N] ([%]) | Positive | Invest |
| [theme] | [N] ([%]) | Mixed | Monitor |
### Top Insight
[Finding] — [Implication]
### What Users Love (Protect This)
[Theme with highest positive sentiment — do not degrade this in future changes]
### Critical Fix Needed
[Theme with highest negative volume and severity]
### Patterns Worth Investigating
[Themes where the signal is interesting but unclear — need more data]
```
## Delivery
If output exceeds the 40-line CLI budget, invoke `/atlas-report` with the full findings. The HTML report is the output. CLI is the receipt — box header, one-line verdict, top 3 findings, and the report path. Never dump analysis to CLI.
当前收录版本
0.6.4
质量评测
满分 100。前四项来自 DeepSkill 收录的质量报告。
评分依据
覆盖多源反馈(工单/NPS/应用商店/流失调研),解决产品团队核心痛点
评分依据
触发词覆盖典型用户表达,边界清晰,无歧义
评分依据
六步分析法逻辑清晰,含具体判断标准和输出模板
评分依据
使用标准平台工具,输入兼容性强
计算方式
取前四项原始分值的算术平均,四舍五入取整。可通过收录报告核对各项评分与说明,运行记录需单独查证。
权限与使用前检查
根据当前主文档自动整理,命中内容可能是示例或防护说明;不是实际权限声明,也未覆盖附件、外部依赖或运行行为。
未从主文档识别到上述常见操作,不能据此判断无需权限或没有风险。
安全评级与访问范围
risk · 静态证据核验,不代表运行安全保证。
使用前请由宿主核对并授权实际文件读写、外部服务和联网目的地;本站的核验不会授予任何运行权限。
用户讨论
还没有已核实的点评。