Canonry: Open-Source AEO Platform for AI Agents

What changed

Canonry has been released as an open-source, agent-first platform for Answer Engine Optimization (AEO). This platform is designed to be self-hosted and offers a comprehensive suite of tools including a command-line interface (CLI), an API, and a web UI. It allows users to monitor AI visibility over time, investigate changes in search and traffic data, and measure the effectiveness of actions taken by agents or human operators. The system utilizes SQLite for data storage and supports Bring Your Own Key (BYOK) for provider keys, ensuring data privacy and control. Both the web UI and agents access the same underlying evidence, promoting consistency.

Getting Started

Users can initiate Canonry through several methods. An agent can be set up by providing a first-time setup request, with guidance available through official Canonry documentation for agent quickstarts, CLI references, and plugin/MCP setup. Alternatively, users can install Canonry locally using npm (npm install -g @canonry/canonry), initialize it with cnry bootstrap, and then serve it locally via cnry serve. After starting the server, users can access a setup page at http://127.0.0.1:4100/setup to scan their site. Initial scans capture a Page Health baseline, with AI Visibility being an optional, separate setup.

Core Functionality

Once running, users can create projects via the CLI, specifying domain, country, and language. Scans can be initiated using commands like cnry technical-aeo run my-site --max-pages 100 --wait --format json. Results, including scores and page-level data, can be retrieved using commands such as cnry technical-aeo score my-site --run-id --format json and cnry technical-aeo pages my-site --run-id --sort score-asc --limit 10 --format jsonl. The platform also supports tracking brand mentions and citations across different answer engines and queries. This involves configuring AI providers like Gemini, OpenAI, Claude, Perplexity, or local models by setting environment variables for API keys. Users can add specific queries and run visibility checks to analyze their share of mentions and citations.

Data Analysis and Actions

Canonry provides tools to investigate visibility changes using data from search performance, crawler visits, AI fetches, referrals, and conversions. Site audits can be performed to identify technical issues alongside performance evidence. The platform supports organizing measurements into portfolios, properties, and market groups for comparative analysis. Agents can leverage Canonry's capabilities to improve sites by integrating with content, publishing, JSON-LD, and indexing workflows. Post-change analysis is facilitated by configuring independent schedules for data synchronization and checks, allowing for direct comparison of new results against historical evidence. Reporting options include generating HTML reports or JSON evidence, with webhook alerts available for integrations like Discord or Slack.

Integrations and Self-Hosting

Canonry boasts a wide range of integrations, covering AI answers (ChatGPT, Claude, Gemini, Perplexity, local models), search and local performance (Google Search Console, Bing Webmaster Tools, Google Business Profile), analytics (Google Analytics 4), conversion measurement, server traffic (Cloudflare, Vercel), backlinks (Common Crawl), publishing (WordPress, JSON-LD), and ChatGPT Ads. The platform is designed for single-tenant self-hosting, with a CLI and REST API serving as primary interfaces. An OpenAPI specification is available at GET /api/v1/openapi.json.

Why it matters for builders

Canonry empowers AI builders by providing a dedicated platform to understand and optimize their presence within the evolving landscape of AI-driven search and answer engines. It offers concrete tools to measure visibility, analyze performance metrics, and directly assess the impact of AI agent actions on search rankings and traffic. This allows developers to move beyond guesswork and implement data-driven strategies for AEO.

Practical impact

Builders can leverage Canonry to set up automated monitoring of their site's AI visibility and search performance. They can integrate their AI agents with the platform to perform AEO tasks, such as content optimization or technical SEO improvements, and then use Canonry to measure the results. The platform's extensive integrations allow for a holistic view of performance, connecting AI visibility data with traffic, conversion, and technical health metrics. Developers can start by installing the CLI and running initial scans to establish a baseline, then explore AI visibility tracking with their preferred LLM providers.

Caveats and source limits

The provided source is a GitHub repository description and documentation. Specific details regarding pricing, independent benchmark results, or a definitive release date beyond "fresh release" are not available. The platform's performance and capabilities are described through its features and intended use cases, but real-world performance metrics or user testimonials are absent. The source focuses on the technical setup and functionality, with less emphasis on user experience beyond the setup process.