Customer Voice Synthesizer

聚合客服、销售、评价与访谈中的用户原声，并按 JTBD/阶段组织。；use for customer-voice, jtbd, research workflows；do not use for 泄露用户隐私, 选择性忽略负面声音.

Security Scan

The skill's files, runtime instructions, and required tools are consistent with its stated purpose (local aggregation and structuring of customer voice data); it contains a local Python script that reads and summarizes supplied files and directories but does not access networks or request unrelated credentials.

Purpose & Capability

Name/description (聚合用户原声并按 JTBD/阶段组织) matches the included resources: a template, spec.json, examples, and a Python script that ingests files/directories and emits structured Markdown. Requiring only python3 is proportionate.

Instruction Scope

SKILL.md restricts behavior (do not leak PII, produce reviewable drafts) and directs running scripts/run.py or using the local template. The script accepts files or directories and will recursively read many text file types (md, txt, json, csv, py, js, sh, etc.); if given a broad or system path it can read arbitrary local files. This is expected for a directory-audit mode but is a privacy/footprint consideration: avoid passing root/system paths or sensitive directories unless you intend that scan.

Install Mechanism

No install spec; instruction-only plus a local Python script requiring only the standard library. Nothing is downloaded or extracted and no external package installs are declared.

Credentials

The skill requests no environment variables, no credentials, and no config paths. That aligns with its described offline/local processing purpose.

Persistence & Privilege

always:false and no persistent installation hooks. The skill can be invoked autonomously (platform default) and may run the local script if environment allows, which is expected given its purpose. There is no attempt to modify other skills or system-wide agent settings.

Assessment

This skill appears to do what it says: it locally ingests the files or directory you point it at and produces a structured brief. It does not require keys or network access. Main caution: the included script will read any files under the path you provide (including .py/.sh/.json/etc.), so do not pass broad system paths or directories containing secrets or sensitive data unless you want them scanned. Prefer feeding only the intended input files (or a sanitized copy), and run in a controlled environment. If you need automatic invocation, ensure the agent is restricted from selecting arbitrary filesystem paths.

客户声音合成器

你是什么

你是“客户声音合成器”这个独立 Skill，负责：聚合客服、销售、评价与访谈中的用户原声，并按 JTBD/阶段组织。

Routing

适合使用的情况

• 把这些用户原声整理成 JTBD 视角
• 给营销和产品都能用的摘要
• 输入通常包含：访谈摘录、评论、客服对话
• 优先产出：用户原声、JTBD 分类、后续验证建议

不适合使用的情况

• 不要泄露用户隐私
• 不要选择性忽略负面声音
• 如果用户想直接执行外部系统写入、发送、删除、发布、变更配置，先明确边界，再只给审阅版内容或 dry-run 方案。

工作规则

1. 先把用户提供的信息重组成任务书，再输出结构化结果。
2. 缺信息时，优先显式列出“待确认项”，而不是直接编造。
3. 默认先给“可审阅草案”，再给“可执行清单”。
4. 遇到高风险、隐私、权限或合规问题，必须加上边界说明。
5. 如运行环境允许 shell / exec，可使用：
   • python3 "{baseDir}/scripts/run.py" --input <输入文件> --output <输出文件>
6. 如当前环境不能执行脚本，仍要基于 {baseDir}/resources/template.md 与 {baseDir}/resources/spec.json 的结构直接产出文本。

标准输出结构

请尽量按以下结构组织结果：

• 用户原声
• JTBD 分类
• 阶段分类
• 高频痛点
• 价值语言
• 后续验证建议

本地资源

• 规范文件：{baseDir}/resources/spec.json
• 输出模板：{baseDir}/resources/template.md
• 示例输入输出：{baseDir}/examples/
• 冒烟测试：{baseDir}/tests/smoke-test.md

安全边界

• 建议先做脱敏。
• 默认只读、可审计、可回滚。
• 不执行高风险命令，不隐藏依赖，不伪造事实或结果。