# Canonry — AI search visibility tracking plus AEO delivery, as an open-source platform, agency products and managed service. > Source: CitedIndex — https://citedindex.com/canonry (structured, researched, re-verified) > Facts last verified: 2026-10-03 Canonry is an answer-engine-optimization company that tracks whether AI assistants name and cite a brand and does the fixing work. It ships a self-hosted open-source platform, agency reporting products, per-market deployments and a managed AEO service. | Fact | Value | | --- | --- | | Website | https://canonry.ai | | Price | Embedded has a price per client. We help you get started and give you an estimate based on your client count. Prices for Custom, Managed, and the full AI Visibility Report depend on the work you need. | | Free tier | Yes | Free page-level site check with no signup, capped at 10 checks per hour from one IP address; the full open-source Canonry platform is MIT-licensed and self-hostable at no licence fee, with provider and infrastructure costs borne by the operator. ## What it does **Monitors AI answers** · Yes — Scheduled runs record mentions, citations and omissions per question and engine, and the same record is what both the service and the self-hosted platform report on. > Scheduled checks record whether each AI tool mentions your brand, cites your site, or leaves you out, along with the competitors it names. > — https://canonry.ai/ · FAQ, 'How do you measure AI visibility?' answer **Optimises your content** · Yes — A free 16-check page audit scores existing pages, returns evidence per check and ranks the top three fixes, and the same model drives paid engagements. > You get a score from 0 to 100, evidence for each check, and the text AI found. The top three fixes are ranked by their likely impact. > — https://canonry.ai/audit · 'How it works' step 4, 'Get your score and top fixes' **Generates content** · Yes — Content writing is part of the AEO work Canonry's own team performs, alongside structured data and brand-fact consistency; it is a service output, not a generator you drive yourself. > We clarify business details, write useful content, add structured data, and keep your brand information consistent across the web. > — https://canonry.ai/ · FAQ, 'What is AEO?' answer **Publishes to your site** · Partly — With integrations configured the platform can push fixes into WordPress and submit pages for indexing, but publishing is documented for WordPress only and on the platform you host yourself. > With the right integrations configured, Canonry can help execute fixes through WordPress, JSON-LD schema workflows, and indexing submissions. > — https://canonry.ai/llms-full.txt · Platform execution-layer description **Technical AEO** · Yes — Technical AEO is the core of the published methodology: JSON-LD carries the heaviest weight, with separate checks for llms.txt, robots.txt, sitemaps and AI bot access. > These 16 checks cover your site's technical foundations, with the most weight given to JSON-LD structured data. > — https://canonry.ai/aeo-methodology · 'What We Score' section intro **Done-for-you service** · Yes — Canonry's own team runs research, fixes, tracking and client reporting under an agency's brand, with the agency keeping pricing and the client relationship. > We deliver AEO under your brand while you keep the client relationship. Our team handles research, planning, technical and content fixes, tracking, and reports > — https://canonry.ai/white-label-aeo-for-agencies · FAQ, 'What is white-label AEO?' answer ## Answer engines **ChatGPT** · Yes — ChatGPT is one of the five engines the vendor says it focuses on, and the self-hosted platform lists OpenAI among its supported providers. > We focus on ChatGPT, Claude, Gemini, Perplexity, and Copilot, and also track Grok, Meta AI, and DeepSeek. > — https://canonry.ai/ · FAQ, 'Which AI tools do you track?' answer **Perplexity** · Yes — Perplexity is named in the focus list and is also a supported provider in the open-source platform, so it is covered on both the service and self-hosted paths. > Canonry supports Gemini, OpenAI, Claude, Perplexity, and local models with OpenAI-compatible APIs. > — https://canonry.ai/open-source-platform · FAQ, 'Which AI tools can it track?' answer **Gemini** · Yes — Gemini is in the vendor's focus list and in the open-source platform's supported provider list, and case-study observations cite Gemini inline citations. > Canonry supports Gemini, OpenAI, Claude, Perplexity, and local models with OpenAI-compatible APIs. > — https://canonry.ai/open-source-platform · FAQ, 'Which AI tools can it track?' answer **Google AI Overviews** · No — The vendor's complete inventory of tracked AI tools names eight chat assistants and no Google search surface; AI Overviews appear only as discussion in blog and methodology prose. > We focus on ChatGPT, Claude, Gemini, Perplexity, and Copilot, and also track Grok, Meta AI, and DeepSeek. > — https://canonry.ai/ · FAQ, 'Which AI tools do you track?' answer **Google AI Mode** · No — Google AI Mode is absent from the eight-engine tracking inventory; the Google-side data Canonry connects is Search Console and Business Profile, not AI Mode placements. > We focus on ChatGPT, Claude, Gemini, Perplexity, and Copilot, and also track Grok, Meta AI, and DeepSeek. > — https://canonry.ai/ · FAQ, 'Which AI tools do you track?' answer **Claude** · Yes — Claude is in the focus list, a supported provider in the self-hosted platform, and one of the three engines whose answers the sentiment feature classifies. > Canonry supports Gemini, OpenAI, Claude, Perplexity, and local models with OpenAI-compatible APIs. > — https://canonry.ai/open-source-platform · FAQ, 'Which AI tools can it track?' answer **Microsoft Copilot** · Partly — Copilot is named among the five engines the service focuses on, but it is absent from the open-source platform's supported provider list, so self-hosted operators do not get it. > Canonry supports Gemini, OpenAI, Claude, Perplexity, and local models with OpenAI-compatible APIs. > — https://canonry.ai/open-source-platform · FAQ, 'Which AI tools can it track?' answer **Grok** · Partly — Grok is listed as tracked but outside the five focus engines, and the self-hosted platform's provider list does not include xAI, so coverage sits with the serviced offerings. > We focus on ChatGPT, Claude, Gemini, Perplexity, and Copilot, and also track Grok, Meta AI, and DeepSeek. > — https://canonry.ai/ · FAQ, 'Which AI tools do you track?' answer **DeepSeek** · Partly — DeepSeek is in the secondary 'also track' group rather than the focus five, and it is not among the open-source platform's named providers. > We focus on ChatGPT, Claude, Gemini, Perplexity, and Copilot, and also track Grok, Meta AI, and DeepSeek. > — https://canonry.ai/ · FAQ, 'Which AI tools do you track?' answer **Meta AI** · Partly — Meta AI is named only in the secondary tracked group, with no provider adapter listed for the self-hosted platform, so it is service-side coverage rather than product-wide. > We focus on ChatGPT, Claude, Gemini, Perplexity, and Copilot, and also track Grok, Meta AI, and DeepSeek. > — https://canonry.ai/ · FAQ, 'Which AI tools do you track?' answer **Coverage in the vendor’s words** · Yes — The vendor states a two-ring inventory: five focus engines plus three also-tracked ones, with the self-hosted build limited to four providers and local models. > We focus on ChatGPT, Claude, Gemini, Perplexity, and Copilot, and also track Grok, Meta AI, and DeepSeek. > — https://canonry.ai/ · FAQ, 'Which AI tools do you track?' answer ## What it reports **Names the cited sources** · Yes — Citation share is a first-class metric beside mention share, and each stored answer keeps the cited URLs and the sources that supported it. > Citation share is the share that link to the client's domain as a source. > — https://canonry.ai/embedded · FAQ, 'What do mention share and citation share mean?' answer **Competitor tracking** · Yes — Competitors found in a tracked brand's own AI answers are reported beside it, including share of answers, and the client portal view exposes the same comparison. > Yes. It shows competitors in that client's AI answers. You can see who AI names or links to instead of your client. > — https://canonry.ai/embedded · FAQ, 'Can the report show competitors?' answer **Brand sentiment** · Partly — Sentiment scoring exists with per-verdict source quotes across three engines, but the vendor labels it experimental, ships it off by default and requires a third-party TypeSafe key. > Canonry 5.30 adds sentiment scores using Jev from TypeSafe. > — https://canonry.ai/blog/canonry-sentiment-analysis-jev · Opening section, second paragraph **Recommendations** · Yes — Output is prescriptive rather than purely descriptive: the audit ranks three fixes by likely impact, and the platform's loop has an agent plan work for human approval. > You get a score from 0 to 100, evidence for each check, and the text AI found. The top three fixes are ranked by their likely impact. > — https://canonry.ai/audit · 'How it works' step 4, 'Get your score and top fixes' **AI crawler analytics** · Yes — Server logs from supported hosts are read without a browser tag and grouped into AI bot visits, user-requested fetches and clicks arriving from AI links, with verification status kept. > We read server logs from supported hosts without adding a browser tag or changing your site's app code. > — https://canonry.ai/custom · 'Server Visits' section, opening paragraph **Wrong answers about you** · Partly — Monitoring records whether the business is misdescribed, but no dedicated false-claim flag or alert is documented; the vendor attributes hallucination detection to other tools in its own comparison table. > The system tracks whether the business is cited, mentioned, absent, or misdescribed across models over time. > — https://canonry.ai/ai-search-visibility · Section 'How Canonry Handles It', Layer 04 Prompt monitoring ## Plans and signing up **Prompt or credit allowance** · Partly — The self-hosted platform meters nothing: you supply provider keys and pay token and grounding costs. The hosted Embedded product is priced per client with no published prompt allowance. > Canonry does not meter checks, but measurement is not free. The operator supplies provider keys and pays the underlying grounding and token costs. > — https://canonry.ai/blog/canonry-vs-paid-ai-visibility-tools · Section 'The number still needs a denominator' **Projects or brands** · Yes — Several brands or markets are the design centre: Custom keeps each market as its own target under one domain, and the platform manages many clients declaratively. > Canonry keeps each result with its market and shows all markets together. > — https://canonry.ai/custom · 'Track Each Market' section, second paragraph **Countries** · Yes — Location is a configured variable rather than an artefact of where the probe runs, so the same prompt set can be run against different markets and each result keeps its location. > you can run the same prompt set from different geographies > — https://canonry.ai/llms-full.txt · Local/geography section on controlled location context **How often it checks** · Yes — The stated service cadence is weekly with diffs, and the self-hosted platform supports both on-demand runs and scheduled runs set by the operator. > Canonry tracks each signal above against each model on a weekly cadence with diffs > — https://canonry.ai/llms-full.txt · 'Ongoing monitoring' section **Card required to start** · No — The free page check asks for nothing at all: no signup, no email and no card, with a published fair-use ceiling instead of a payment wall. > Full results are free, with no signup. > — https://canonry.ai/audit · Hero paragraph under 'One page. 16 checks. Three fixes.' **Self-serve signup** · Partly — The free audit and the self-hosted platform need no contact at all, but every paid offering (Embedded, Custom, Managed, the full report) is scoped and priced through Canonry first. > Embedded has a price per client. We help you get started and give you an estimate based on your client count. > — https://canonry.ai/ · FAQ, 'What does AEO cost?' answer ## Access and integrations **API** · Yes — The self-hosted platform exposes a versioned REST API alongside CLI and webhooks, with the dashboard described as just another client of the same project API. > Other agents can use CLI commands with JSON output, a versioned REST API, or MCP tools loaded as needed. > — https://canonry.ai/open-source-platform · FAQ, 'How does my agent connect?' answer **MCP server** · Yes — An MCP adapter fronts the same project API for any MCP-capable agent, with native Codex and Claude Code plugins as the alternative path (the vendor says not to run both). > Canonry gives your agent evidence, tools, spending limits, and approval checks through CLI, REST, MCP, webhooks, or a Codex or Claude Code plugin. > — https://canonry.ai/open-source-platform · FAQ, 'Does Canonry replace my agent?' answer **BI export** · Partly — No Looker, Power BI or warehouse connector is published; the export path is the operator-owned SQLite file plus REST and JSON CLI output. > By default, its project data lives in one SQLite file on your machine or server. You control the provider keys, backups, and deployment. > — https://canonry.ai/open-source-platform · Direct answers FAQ, 'Where does Canonry store my data?' ## Repository (as of 2026-08-30) | Fact | Value | | --- | --- | | Stars | 126 | | Forks | 14 | | Last push | 2026-08-30 | ## CitedIndex in AI answers 1 of 4 answers cite CitedIndex. | Engine | Result | Question | Asked on | Other sources | | --- | --- | --- | --- | --- | | Perplexity | Cites CitedIndex | Canonry alternatives | 2026-10-12 | canonry.ai, libhunt.com, agentskillshub.top, openalternative.co, reddit.com, academicdictionary.com | | ChatGPT | Does not cite CitedIndex | Canonry alternatives | 2026-10-12 | producthunt.com, learnist.org, app.rankology.io, tooliverse.ai, toolhunt.eu, tryanalyze.ai | | Claude | Listed no sources | Canonry alternatives | 2026-09-04 | — | | Gemini | Does not cite CitedIndex | Canonry alternatives | 2026-09-04 | encyclopedia.pub, wikipedia.org, thepriest.com, globalsistersreport.org, futurechurch.org, vatican.va | ## What changed - 2026-10-03 — Capabilities · Free tier · Price - 2026-09-02 — Capabilities - 2026-09-01 — Capabilities · Free tier - 2026-08-31 — Capabilities · Free tier - 2026-08-30 — Capabilities · Links - 2026-08-22 — Alternatives · Summary · Tagline ## Corrections If you represent Canonry, or anything here is wrong or out of date, send a correction with a source. The operations this site accepts from an agent are listed at https://citedindex.com/.well-known/agents.json — that catalog needs no key to read, and each entry states its own auth.