hearsay-ai-visibility

仓库创建 2026年7月27日最近提交 2 天前SkillHot 收录 2 天前

▸ 精选理由

提供跨模型统计和置信区间，适合做品牌在 AI 中被引用的定量监测。

▸ 风险提示

需启动本地 Node 服务并可能上传/收集品牌相关数据。

这个 Skill 做什么

操作本地 Hearsay 跟踪器以测量在各大 LLM 中的品牌与 AI 语音份额。

在本地用 Hearsay 跟踪器把品牌在各大 LLM（ChatGPT、Claude、Gemini、Perplexity）里的“AI 语音”和提及量做量化监测。遇到需要做 AI visibility、AI SEO、地域（GEO）差异分析或比对不同模型表现时用它。特别之处是输出可核查的统计数据（置信区间、样本量、措辞分布），方便做严谨的报告与决策。

▸ 展开 SKILL.md 英文原文

Operate and interpret a local Hearsay AI-visibility tracker via its MCP tools (hearsay_*) — set up brand tracking, generate prompt panels, run measurements, and report share of AI voice / brand mentions in ChatGPT, Claude, Gemini and Perplexity with honest statistics (confidence intervals, n, phrasing spread). Use for AI visibility, GEO, AI SEO, LLM brand monitoring questions.

数据与抓取品牌监测LLM可见性本地工具通用

Stars

Forks

仓库内 Skill

7 日增星

安装 / 使用

给你的 Agent 一句话（通用）

帮我安装这个 skill：https://raw.githubusercontent.com/melonwer/hearsay/main/skill/SKILL.md

或 curl 直取 SKILL.md

curl -fsSL "https://raw.githubusercontent.com/melonwer/hearsay/main/skill/SKILL.md"

SKILL.MD 节选查看完整文件 ↗

# Operating Hearsay

## First move — always

Call `hearsay_status` before anything else, then branch:

- **Unreachable** → tell the human to start it: `node server.js` in the hearsay
  directory (needs Node ≥ 22.13), then retry.
- **`configured: false`** → run the onboarding playbook below.
- **Configured** → answer the question with the *narrowest* tool that holds the
  answer (summary for headlines, intent_results for per-question, answers_search
  for receipts). Don't sweep every tool.

## Onboarding playbook (unconfigured instance)

1. Ask the human at most two questions: their brand (+ domain), and their main
   competitors. Ask for a category description only if it isn't obvious.
2. `h

via SKILL·HOT · 数据来自 GitHub 公开信息 · 原文版权归作者所有