Dageno Agent Project Map / Dageno Agent 项目导航

Docs / 文档: README · 简体中文

If this repo is useful, you may also want the adjacent Dageno Agent projects for GEO, SEO, AI visibility, and content operations. 如果这个仓库对你有帮助，也可以看看这些相邻的 Dageno Agent 项目，用于 GEO、SEO、AI 可见性和内容增长工作流。

If you want to... / 如果你想...	Project / 项目	Plain-language difference / 白话区别
First diagnose a site / 先给网站做体检	seo-geo-audit	Like an SEO + GEO medical report: technical issues, content gaps, trust signals, off-site mentions, and AI visibility in one audit / 像一份 SEO + GEO 体检报告，把技术问题、内容缺口、信任信号、站外提及和 AI 可见性放到一起看
Turn a real website into Dageno topics and prompts / 把真实网站变成 Dageno 监控题库	dageno-online-topic-prompt-generator	Crawls the site and studies the business first, then generates Topic clusters and high-intent Prompts. Not an industry-template prompt dump / 先看网站和业务，再生成 Topic 集群和高意图 Prompt，不是套行业模板
Produce SEO/GEO articles from keywords or briefs / 从关键词或 brief 批量生产内容	seo-geo-content-engine	A full content pipeline: research, SERP intent, article structure, draft, metadata, FAQ, and GEO packaging / 完整内容流水线：调研、搜索意图、文章结构、正文、metadata、FAQ 和 GEO 包装
Write from Dageno fanout data / 用 Dageno fanout 写文章	geo-content-writer	For when Dageno already found prompt opportunities: turn fanout into a backlog, editorial brief, draft contract, and review contract / 适合已经有 Dageno prompt opportunity 的情况：把 fanout 变成选题队列、编辑 brief、草稿契约和审核契约
Find why organic content is not converting / 找出自然流量内容为什么不转化	organic-content-intelligence	Joins GSC, GA4, crawl, intent, and AI/GEO signals to show which pages have demand but fail to answer or convert / 把 GSC、GA4、抓取、意图和 AI/GEO 信号连起来，看哪些页面有需求但没有承接住
Improve a site's GEO structure / 优化网站结构以适配 GEO	geo-site-architecture-audit	Starts from the existing navigation, sitemap, landing pages, and help content, then finds missing AI-answerable pages and internal links / 从现有导航、站点地图、落地页和帮助内容出发，找缺失的 AI 可引用页面和内链结构
Create a client-facing AI visibility report / 做给客户看的 AI 可见性报告	brand-ai-performance-check	A stable visual report template for brand AI performance, using Dageno API data or custom inputs / 稳定的品牌 AI 表现可视化报告模板，可接 Dageno API 或自定义数据
Automate Dageno in workflows / 把 Dageno 接进自动化流程	n8n-nodes-dageno	Use Dageno inside n8n: brands, GEO analysis, keywords, opportunities, topics, prompts, SEO, and citations / 在 n8n 里调用 Dageno：品牌、GEO 分析、关键词、机会、Topic、Prompt、SEO 和引用数据
Learn the API and MCP growth workflow / 学 Dageno API 和 MCP 怎么用于增长	dageno-mcp-growth-playbook	The practical playbook for turning Dageno API/MCP data into reports, prompt gaps, citation intelligence, and growth actions / 把 Dageno API/MCP 数据变成报告、Prompt Gap、引用分析和增长动作的实战手册

More projects / 更多项目: geo-visual-content-engine, seo-outreach-skill, geo-pre-sale-report-private, GEO-SEO.

Explore all repos / 查看全部项目: github.com/dageno-agents · Product / 产品: Dageno

Organic Content Intelligence

Open-source diagnostics for organic content growth: search demand, page funnels, intent coverage, data quality, AI/GEO visibility, and optimization task drafts.

Organic Content Intelligence helps teams answer a harder question than "which page lost traffic?"

It asks:

Is this content actually capturing the demand it attracts?

Most SEO dashboards stop at impressions, clicks, CTR, and Position. Most analytics dashboards start after the user lands on the page. This project joins the two sides, then adds intent evidence, content coverage, freshness checks, AI/GEO signals, and data quality guardrails.

The result is a portable framework for diagnosing where organic content fails to carry users from search demand to useful action.

Why This Exists

Content teams often have data, but not a diagnosis.

Search Console can show that a page gets impressions for new queries. GA4 can show that the page has weak engagement or few CTA clicks. A crawler can show that AI bots visited the page. A content audit can show that pricing, screenshots, or product claims are stale.

But those signals usually live in different tools.

Organic Content Intelligence turns them into one workflow:

flowchart LR
  A["Search demand<br/>GSC queries + pages"] --> B["Intent clusters<br/>What users want"]
  C["Page behavior<br/>GA4 page performance"] --> D["Page funnel<br/>Where users drop"]
  E["Content crawl<br/>Headings, FAQs, CTAs, snippets"] --> F["Coverage map<br/>Where intent is answered"]
  G["Brand knowledge<br/>Pricing, features, screenshots, FAQ"] --> H["Freshness checks<br/>What is stale or unsafe"]
  I["AI/GEO signals<br/>AI referrals + crawler logs"] --> J["Visibility checks<br/>Can AI systems reach it"]
  B --> K["Diagnosis"]
  D --> K
  F --> K
  H --> K
  J --> K
  K --> L["Data quality gate"]
  L --> M["Optimization task draft"]

Loading

What It Can Diagnose

Problem	What the system checks	Example output
Traffic risk	GSC impressions, clicks, CTR / CTR Δ, Position movement	"This page has high exposure but worsening rank and CTR."
Conversion weakness	GA4 page-level sessions, engagement, CTA clicks, key events	"The page receives organic sessions but weak CTA interaction."
Intent gap	Query clusters vs page sections, FAQs, tables, CTAs	"New tool-intent queries have impressions, but no workflow section exists."
Stale content	Brand knowledge base vs article claims, screenshots, pricing, FAQ	"Pricing copy is outdated against knowledge base version 2026-06."
Cannibalization	Multiple pages competing for the same intent cluster	"Two articles split the same commercial intent."
AI/GEO visibility	AI referral sessions and crawler logs	"AI referrals exist, but ClaudeBot is blocked on this URL."
Data risk	Missing, incomplete, or low-confidence data	"Do not generate a task yet; manual verification required."

Product Surface

The demo contains five surfaces. They are deliberately generic, so teams can adapt them to their own stack.

flowchart TD
  O["1. Page Funnel Overview"] --> I["2. Issue Groups"]
  I --> D["3. Single Page Diagnosis"]
  D --> T["Task Draft"]
  G["4. AI/GEO Signals"] --> D
  C["5. Intent Cluster Overview"] --> D
  Q["Data Quality"] --> O
  Q --> I
  Q --> D
  Q --> C
  K["Brand Knowledge Base"] --> D

Loading

1. Page Funnel Overview

The page-level command center.

It combines:

• GSC: impressions, clicks, CTR / CTR Δ, Position
• GA4-style page performance: organic sessions, engagement rate, CTA clicks, key events
• Risk scores: Traffic Risk Score, Conversion Weak Score
• Data Quality: whether the system can safely recommend action

The page list is not the final task generator. It is the entry point into diagnosis.

2. Issue Groups

Groups pages by problem type:

• CTR decline
• Ranking or impressions decline
• Conversion weakness
• Stale content
• New query opportunity
• Intent cannibalization
• AI/GEO visibility weakness

This helps teams work in batches instead of inspecting URLs one by one.

3. Single Page Diagnosis

The evidence room for one page.

It shows which query clusters the page attracts, what intent each cluster represents, how well the page covers that intent, and where the content should be improved.

Query-to-content mapping is shown as an expandable detail layer, not a permanent table. This keeps the page readable while preserving evidence.

flowchart LR
  A["Query cluster<br/>track brand mentions in ChatGPT"] --> B["Intent<br/>Tool / workflow"]
  B --> C["Current page match<br/>Short paragraph"]
  C --> D["Gap<br/>Coverage is shallow"]
  D --> E["Recommended module<br/>New H2 + checklist + CTA"]

Loading

4. AI / GEO Signals

Two related but separate tables:

• AI Referral Sessions from analytics
• AI Crawler Logs from server, edge, or CDN logs

The project keeps these signals separate because referral traffic and crawler accessibility answer different questions.

5. Intent Cluster Overview

A library-level view.

Instead of asking "which page is weak?", this view asks "which user intents are we covering well or poorly across the whole content library?"

This is useful for content strategy, pruning, consolidation, and roadmap planning.

Data Model at a Glance

erDiagram
  PAGES ||--o{ GSC_PAGE_DAILY : has
  PAGES ||--o{ GA4_PAGE_DAILY : has
  PAGES ||--o{ QUERY_CONTENT_MAPPING : contains
  QUERY_CLUSTERS ||--o{ QUERY_CLUSTER_MEMBERS : groups
  QUERY_CLUSTERS ||--o{ QUERY_CONTENT_MAPPING : explains
  PAGES ||--o{ DATA_QUALITY_CHECKS : checked_by
  BRAND_KB_SOURCES ||--o{ BRAND_KB_ITEMS : provides
  PAGES ||--o{ FRESHNESS_CHECKS : evaluated_by
  QUERY_CLUSTERS ||--o{ INTENT_CLUSTER_SUMMARY : aggregates

Loading

Data Quality Gate

Every recommendation reads from the same object:

{
  "quality_grade": "High",
  "reason_codes": [],
  "checked_at": "2026-06-16T00:00:00Z"
}

Grade	Meaning	UI action
High	Evidence is good enough for automation	Generate task draft
Medium	Useful, but needs human verification	Needs verification
Low	Useful for diagnosis only	Diagnostics only
Invalid	Do not judge	No judgment

This guardrail matters because content optimization systems can easily look more certain than the data allows.

Important Attribution Boundary

GA4 data is treated as page-level performance.

Query clusters explain search intent and content fit. They do not prove that a specific query cluster directly caused a conversion.

Use:

GA4 Page Performance, not query attribution

Allowed attribution labels:

• estimated
• directional

Quick Start

This seed project has no build step and no external dependencies.

git clone https://github.com/dageno-agents/organic-content-intelligence.git
cd organic-content-intelligence
npm run dev

Open:

http://localhost:4173/public/

Run fixture checks:

npm run check

For Humans and Agents

If you are evaluating the project:

• Start with this README.
• Read Page Walkthrough to understand the five UI surfaces.
• Read Data and API Spec to connect GSC, GA4, crawler logs, and brand knowledge.
• Read Architecture to understand the pipeline.

If you are a coding agent:

• Start with Agent Guide.
• Do not invent query-level conversion attribution.
• Keep Data Quality shared across all surfaces.
• Keep private customer data, internal prompts, and proprietary weights out of the open-source core.

Repository Structure

organic-content-intelligence/
  public/                 Static demo entry
  src/                    Mock data, UI rendering, scoring, URL normalization
  schemas/                JSON Schemas for shared contracts
  examples/               GSC and GA4 request examples
  docs/                   Architecture, data spec, page walkthrough, roadmap
  scripts/                Fixture validation

Open-Core Boundary

Good open-source core:

• Dashboard shell
• Mock data and schemas
• GSC / GA4 adapter contracts
• URL normalization
• Data Quality framework
• Basic scoring examples
• Query cluster and intent overview models

Better kept in a hosted or private layer:

• Customer data
• Proprietary scoring weights
• Internal LLM prompts
• Private brand knowledge base content
• Publishing workflow and approvals
• Managed sync infrastructure

Naming Note

The original prototype used the slug seo-content-funnel, but the project direction is broader than an SEO funnel. The public product name is Organic Content Intelligence because it covers organic search demand, content coverage, page performance, freshness, AI/GEO visibility, and task drafting.

References

• GitHub README guidance: https://docs.github.com/en/repositories/managing-your-repositorys-settings-and-features/customizing-your-repository/about-readmes
• GitHub Mermaid diagrams: https://docs.github.com/en/get-started/writing-on-github/working-with-advanced-formatting/creating-diagrams
• Diataxis documentation framework: https://diataxis.fr/

License

MIT