Adology AI
Perplexity for social: ask what competitors are running across social and get answers grounded in real ad and creative data, not guesses.
README
Adology — Content Intelligence
Competitive ad intelligence inside Claude and other AI agents. Discover brands, build knowledge sets, analyze ad content, coach creators, vet influencers, run audience deep-dives, and generate strategic insights — all powered by Adology's data platform.
What This Plugin Does
The content-intelligence plugin connects your AI agent to your Adology workspace, giving you access to competitive ad intelligence directly in conversation. You can search for brands, manage knowledge sets, run content analysis, coach influencer accounts, vet creator partnerships, and produce strategy artifacts without leaving your AI agent.
Capabilities
- Knowledge Set Management — Create, configure, and manage knowledge sets that track brands and competitors across ad platforms.
- Brand Discovery — Search for brands across TikTok, Instagram, Facebook, YouTube, and more.
- Competitive Analysis — Compare ad strategies, creative approaches, and content patterns across brands.
- Content Intelligence — Analyze ad content with AI-powered labeling, structural analysis, and pattern detection.
- Creative Toolkits — Build living libraries of hooks, angles, and scripts grounded in competitor + audience data, in the brand's actual voice.
- Influencer Coaching & Vetting — Coach creator accounts with personalized feedback, and score/vet influencer partners with evidence-based fit analysis.
- Audience Deep Dives — Produce behaviorally grounded audience profiles (empathy canvas, insight brief) backed by social + Reddit data.
- Brand Marketing Mode — Strategic brand thinking across positioning, competitive landscape, content strategy, and go-to-market.
- Data Export — Export filtered datasets, reports, and analysis results.
What's Included
| Component | Count | Description |
|---|---|---|
| MCP Tools | 27 | Server-side tools for data access and operations |
| Domain Skills | 11 | Guided workflows for common analysis and strategy tasks |
| Slash Commands | 4 | Quick-access commands for frequent actions |
Skills
| Skill | Purpose |
|---|---|
getting-started |
Onboarding walkthrough for first-time users |
brand-builder |
Build a knowledge set around a brand and its competitors |
content-strategist |
Quick creative analysis: hooks, formats, CTAs |
data-explorer |
Query and explore Adology data fields directly |
research-analyst |
Multi-step research patterns; standard and heavy modes |
brand-marketing-mode |
Strategic brand thinking: positioning, landscape, content, go-to-market |
audience-deep-dive |
Behaviorally grounded audience profile + empathy canvas + insight brief |
creative-toolkit |
Living library of hooks/angles/scripts in the brand's voice, data-sourced |
influencer-coach |
Personalized creator coaching from knowledge set + account analysis |
influencer-vetting |
Score and select influencer/creator partners with evidence-based fit analysis |
thumbnails |
Reliable embedding of Adology thumbnails in generated documents |
Installation
See INSTALL.md for a full step-by-step guide covering the Claude.ai connector, Claude Code plugin, and Claude Cowork plugin install paths.
Quick links:
- Claude.ai — add
https://mcp.adologyai.com/mcpas a custom connector under Settings → Connectors and authenticate via Stytch. - Claude Code — inside a session, run
/plugin marketplace add adologyai/content-intelligence-pluginfollowed by/plugin install content-intelligence@adology-marketplace. No clone required. - Claude Cowork — in Claude Desktop, open the Cowork tab → Customize → Browse plugins and install the Adology plugin (or upload the latest release zip until the directory listing is live).
Account Setup
You need an Adology account to use this plugin. Sign up at https://adologyai.com. The plugin uses OAuth 2.1 (Stytch) — your AI agent will prompt you to sign in the first time it calls an Adology tool. No API token to manage.
License
Apache 2.0 — see LICENSE.
The plugin code is freely usable under Apache 2.0. Access to the Adology service is governed by the Adology Terms of Service.
Documentation
INSTALL.md— customer install guide (Claude.ai + Claude Code)docs/installation-claude-code.md— Claude Code install referencedocs/installation-connector.md— Claude.ai connector referencedocs/tools-reference.md— MCP tools referencedocs/skills-reference.md— bundled skills reference
Contact
- General: hello@getadology.com
- Privacy: see the Privacy Policy
- Issues: https://github.com/adologyai/content-intelligence-plugin/issues
推荐服务器
Baidu Map
百度地图核心API现已全面兼容MCP协议,是国内首家兼容MCP协议的地图服务商。
Playwright MCP Server
一个模型上下文协议服务器,它使大型语言模型能够通过结构化的可访问性快照与网页进行交互,而无需视觉模型或屏幕截图。
Magic Component Platform (MCP)
一个由人工智能驱动的工具,可以从自然语言描述生成现代化的用户界面组件,并与流行的集成开发环境(IDE)集成,从而简化用户界面开发流程。
Audiense Insights MCP Server
通过模型上下文协议启用与 Audiense Insights 账户的交互,从而促进营销洞察和受众数据的提取和分析,包括人口统计信息、行为和影响者互动。
VeyraX
一个单一的 MCP 工具,连接你所有喜爱的工具:Gmail、日历以及其他 40 多个工具。
graphlit-mcp-server
模型上下文协议 (MCP) 服务器实现了 MCP 客户端与 Graphlit 服务之间的集成。 除了网络爬取之外,还可以将任何内容(从 Slack 到 Gmail 再到播客订阅源)导入到 Graphlit 项目中,然后从 MCP 客户端检索相关内容。
Kagi MCP Server
一个 MCP 服务器,集成了 Kagi 搜索功能和 Claude AI,使 Claude 能够在回答需要最新信息的问题时执行实时网络搜索。
e2b-mcp-server
使用 MCP 通过 e2b 运行代码。
Neon MCP Server
用于与 Neon 管理 API 和数据库交互的 MCP 服务器
Exa MCP Server
模型上下文协议(MCP)服务器允许像 Claude 这样的 AI 助手使用 Exa AI 搜索 API 进行网络搜索。这种设置允许 AI 模型以安全和受控的方式获取实时的网络信息。