Skip to content
Back to search

canonry

Official Canonry plugins for Claude Code.

Safety scan

A static text scan, not a guarantee. Select the badge to see what was scanned and every finding.

Checked today

Check it deeper on your machine

It has changed since its scan on 11 Nov 2026.

As a Claude Code plugin, install it and then ask Claude Code:

/plugin marketplace add keithmackay/mackayi/plugin install scanskillsafety@mackayiDo a deep scan of https://github.com/Canonry/canonry with scanskillsafety

Or from a terminal, clone it and run it:

git clone https://github.com/keithmackay/scanskillsafety-skill ~/.claude/skills/scanskillsafetynode ~/.claude/skills/scanskillsafety/cli.cjs --deep 'https://github.com/Canonry/canonry'

It checks the exact commit: every file, its packages against OSV (api.osv.dev), how it compares with this page's scan (findsafeskills.com receives the repository's address), and, if you agree, Claude reading every file with your own Claude subscription. It never runs the code.

Badge for your README
findsafeskills safety rating

Markdown

[![findsafeskills safety rating](/api/listings/canonry--canonry/badge.svg)](/listings/canonry--canonry)

HTML

<a href="/listings/canonry--canonry"><img src="/api/listings/canonry--canonry/badge.svg" alt="findsafeskills safety rating" /></a>

Install

i

Install instructions from the README

As written by the repo's author. findsafeskills has not verified these steps.

<a id="or-use-any-shell-capable-coding-agent"></a>

Start with your agent

<details> <summary>Copy the first-time setup request</summary>

Help me set up Canonry for my public site.

Use the official Canonry docs:
- Agent quickstart: https://github.com/Canonry/canonry#or-use-any-shell-capable-coding-agent
- CLI reference: https://github.com/Canonry/canonry/blob/main/skills/canonry/references/canonry-cli.md
- Plugin setup: https://github.com/Canonry/canonry/blob/main/docs/plugins.md
- MCP setup: https://github.com/Canonry/canonry/blob/main/docs/mcp.md

Pick one path and stay on it. Use connected Canonry tools (plugin or MCP) only if you have them and I have not pointed you at a specific install. A connected tool runs in its own process and never sees `CANONRY_CONFIG_DIR`, `CANONRY_PORT`, or anything else you export in a shell, so it always acts on my default install. If I named a config directory, a port, or a sandbox, that is the shell path: use `cnry` and tell me which path you chose. Never mix the two in one run, and never create a duplicate project. The `cnry` and `canonry` commands are interchangeable.

1. Ask for my public domain, country, and language. Do not create or scan anything yet. Country and language are only applied when a project is created, so if you reuse an existing project, report its country and language instead of changing them.
2. Shell path only: confirm `cnry` is on PATH, then run `cnry --version`. If Canonry is missing, propose `npm install -g @canonry/canonry` and wait for approval. Then run `cnry doctor --format json`, which is the command that says whether config, database, and server are in place. `cnry --version` does not read config and succeeds on a completely unconfigured install, so it cannot answer this. If config is missing, run `cnry bootstrap` yourself: it is not interactive, takes about a second, and is safe to rerun. Do not hand it to me and wait. The interactive command is `cnry init`, which is optional provider and OAuth setup that Page Health does not need. Bootstrap prints an API key, so do not repeat its output back to me, and never ask me to paste passwords, API keys, OAuth credentials, or command output.
3. Confirm the API is reachable. `cnry doctor --format json` reports it, and any project read exits non-zero with `CONNECTION_ERROR` when it is not. If it is unreachable, propose `cnry start` and wait for approval. Use `cnry start`, which is the background daemon, and not `cnry serve`, which runs in the foreground and will block you until I stop it, even though some error messages suggest it. Stop anything you started with `cnry stop`.
4. List projects with the connected project tool or `cnry project list --format json`, and reuse one whose domain matches. Confirm the proposed name is not already assigned to a different domain. To find out whether a project has already been scanned, run `cnry technical-aeo score <project> --format json` with no `--run-id`, which reports the latest run. Read the `hasData` field, not the score: this command exits 0 and reports `aggregateScore: 0` for a project that has never been scanned, so reading the score alone would have you tell me my site scored zero. If `hasData` is true and `runStatus` is `completed` or `partial`, read that scan instead of starting a new one. If no project matches, show the exact create operation and wait for approval.

What was scanned

The files the safety scan read, as written by the repository's author and shown as plain text. Lines with a finding are marked.

.claude-plugin/marketplace.json · Manifest · 50 lines

View this file on GitHub (read at the repository's current version: the scan did not record which commit it read)

This file has changed since it was scanned, so this is not exactly the text that was scanned, and no line is marked. The findings are listed with the safety scan above.

  1. {
  2. "$schema": "https://json.schemastore.org/claude-code-marketplace.json",
  3. "name": "canonry",
  4. "description": "Official Canonry plugins for Claude Code.",
  5. "owner": {
  6. "name": "Canonry"
  7. },
  8. "plugins": [
  9. {
  10. "name": "canonry",
  11. "source": "./plugins/canonry",
  12. "displayName": "Canonry",
  13. "description": "Operate Canonry AEO projects through guided skills and typed MCP tools.",
  14. "author": {
  15. "name": "Canonry"
  16. },
  17. "homepage": "https://canonry.ai",
  18. "repository": "https://github.com/Canonry/canonry",
  19. "license": "MIT",
  20. "keywords": [
  21. "aeo",
  22. "answer-engine-optimization",
  23. "citations",
  24. "mentions",
  25. "mcp"
  26. ],
  27. "category": "productivity"
  28. },
  29. {
  30. "name": "canonry-remote",
  31. "source": "./plugins/canonry-remote",
  32. "displayName": "Canonry Remote",
  33. "description": "Read stored Canonry measurements in a hosted workspace through a read-only OAuth connection.",
  34. "author": {
  35. "name": "Canonry"
  36. },
  37. "homepage": "https://canonry.ai",
  38. "repository": "https://github.com/Canonry/canonry",
  39. "license": "MIT",
  40. "keywords": [
  41. "aeo",
  42. "citations",
  43. "mentions",
  44. "analytics",
  45. "mcp"
  46. ],
  47. "category": "productivity"
  48. }
  49. ]
  50. }
README.md · README · 358 lines

View this file on GitHub (read at the repository's current version: the scan did not record which commit it read)

This file has changed since it was scanned, so this is not exactly the text that was scanned, and no line is marked. The findings are listed with the safety scan above.

  1. <h1 align="center">
  2. <a href="https://canonry.ai"><img src="https://raw.githubusercontent.com/Canonry/canonry/main/apps/web/public/apple-touch-icon.png" alt="Canonry" width="64" valign="middle" /></a> Canonry
  3. </h1>
  4. <p align="center">
  5. <a href="https://www.npmjs.com/package/@canonry/canonry"><img src="https://img.shields.io/npm/v/@canonry/canonry?style=flat&amp;label=npm&amp;color=D93F45" alt="npm version" /></a>
  6. <a href="https://www.npmjs.com/package/@canonry/canonry"><img src="https://img.shields.io/npm/dt/@canonry/canonry?style=flat&amp;label=downloads&amp;color=D93F45" alt="Total npm downloads" /></a>
  7. <a href="./LICENSE"><img src="https://img.shields.io/badge/license-MIT-D93F45?style=flat" alt="License: MIT" /></a>
  8. <a href="https://discord.gg/8K9yY7GywJ"><img src="https://img.shields.io/badge/Discord-5865F2?logo=discord&amp;logoColor=white" alt="Join the Canonry Discord" /></a>
  9. <a href="https://x.com/canonry_ai"><img src="https://img.shields.io/badge/X-000000?logo=x&amp;logoColor=white" alt="Follow Canonry on X" /></a>
  10. <img src="https://img.shields.io/badge/API%20%7C%20CLI%20%7C%20MCP%20%7C%20Web-4493F8?style=flat-square" alt="Interfaces: REST API, CLI, MCP, and web dashboard" />
  11. </p>
  12. <p align="center">
  13. <strong>The agent-first AEO operating platform.</strong><br/>
  14. Track mentions and citations across ChatGPT, Claude, Gemini, Meta Muse, and more. Join them with Search Console, GA4, ads, AI crawler traffic, Agent Readiness site audits, and SEO data. Then let your agent create content, publish fixes, manage ChatGPT ads, get pages indexed, and measure what changed.<br/>
  15. <sub>Open source · Self-hosted · Bring your own keys</sub>
  16. </p>
  17. <h3 align="center"><a href="https://custom-demo.canonry.ai/projects/summit-roofing"><ins>Try the live demo</ins></a> &nbsp;·&nbsp; <a href="#quick-start"><ins>Quick start</ins></a></h3>
  18. <p align="center">
  19. <a href="https://custom-demo.canonry.ai/projects/summit-roofing"><picture><source media="(max-width: 600px)" srcset="https://raw.githubusercontent.com/Canonry/canonry/main/docs/images/readme/hero-mobile.png" /><img src="https://raw.githubusercontent.com/Canonry/canonry/main/docs/images/readme/hero.png" alt="Canonry dashboard: AI answers over time, with the share of non-brand answers that mention the brand for Claude, Gemini, and OpenAI, with sample data" width="100%" /></picture></a>
  20. </p>
  21. <a id="explore-the-evidence"></a>
  22. <a id="add-ai-visibility-when-you-need-it"></a>
  23. ## Features
  24. ### Built for Agents
  25. Everything runs on one REST API. The dashboard, CLI, MCP server, and plugins all call it, so your agent can read and do everything you see (and more!).
  26. - **REST API:** every dashboard number has a typed endpoint under `/api/v1`. The full OpenAPI spec is at `GET /api/v1/openapi.json`. [API →](#self-hosting-and-api)
  27. - **CLI:** the same calls as `cnry <command> --format json`, for scripts and shell agents. [CLI →](skills/canonry/references/canonry-cli.md)
  28. - **MCP and plugins:** plugins for Claude Code and Codex, or a hosted MCP server for ChatGPT and Claude Desktop. [MCP →](docs/mcp.md) · [Plugin →](docs/plugins.md)
  29. ### AI Visibility
  30. Track brand mentions and citations by query and answer engine. See where competitors win, then open the exact answer and sources behind each result. [Docs →](docs/query-visibility.md)
  31. <a href="https://custom-demo.canonry.ai/projects/summit-roofing"><picture><source media="(max-width: 600px)" srcset="https://raw.githubusercontent.com/Canonry/canonry/main/docs/images/readme/ai-visibility-mobile.png" /><img src="https://raw.githubusercontent.com/Canonry/canonry/main/docs/images/readme/ai-visibility.png" alt="Where competitors are winning: mention share against two competitors, mention and citation gaps, and mentioned and cited counts by engine, with sample data" width="100%" /></picture></a>
  32. ### Server-Side AI Traffic
  33. See what AI agents actually do on your site, from your server logs: which pages GPTBot, ClaudeBot, Meta AI, and Grok crawl, which pages ChatGPT, Claude, and Meta AI open for users, and who clicks through. [Docs →](skills/canonry/references/server-side-traffic.md)
  34. <a href="skills/canonry/references/server-side-traffic.md"><picture><source media="(max-width: 600px)" srcset="https://raw.githubusercontent.com/Canonry/canonry/main/docs/images/readme/traffic-mobile.png" /><img src="https://raw.githubusercontent.com/Canonry/canonry/main/docs/images/readme/traffic.png" alt="Server traffic for the last 24 hours: 105 AI crawler visits, 35 AI user fetches, and 46 AI referral sessions, with an hourly chart, with sample data" width="100%" /></picture></a>
  35. <a id="site-health"></a>
  36. ### Agent Readiness
  37. A site audit for AI agents. Crawl your site and score how well agents and answer engines can reach, read, and cite each page: AI crawler access, `llms.txt`, structured data, extractable content, and more. Each finding lists the pages it affects, and a map shows the internal links between pages. [Docs →](skills/canonry/references/canonry-cli.md#agent-readiness) · [Audit engine →](https://github.com/Canonry/agent-readiness)
  38. <a href="skills/canonry/references/canonry-cli.md#agent-readiness"><picture><source media="(max-width: 600px)" srcset="https://raw.githubusercontent.com/Canonry/canonry/main/docs/images/readme/site-health-mobile.png" /><img src="https://raw.githubusercontent.com/Canonry/canonry/main/docs/images/readme/site-health.png" alt="Site map of 205 crawled pages and the links between them, beside a list of site sections, with sample data" width="100%" /></picture></a>
  39. <a id="seo-data"></a>
  40. ### SEO Research
  41. Keyword volume, difficulty, organic rankings, and search gaps against competitors, powered by DataForSEO Labs. [Docs →](docs/seo-data.md)
  42. <a href="docs/seo-data.md"><picture><source media="(max-width: 600px)" srcset="https://raw.githubusercontent.com/Canonry/canonry/main/docs/images/readme/seo-mobile.png" /><img src="https://raw.githubusercontent.com/Canonry/canonry/main/docs/images/readme/seo.png" alt="SEO page with ranked searches, top-10 count, estimated organic traffic, and a table of positions, volume, and difficulty, with sample data" width="100%" /></picture></a>
  43. ### Ads
  44. Let your agent run ChatGPT Ads: sync campaigns and daily results, create and edit campaigns, and switch them on only after you approve. Track Google Ads spend and conversions, and check each conversion's setup across Google Ads and Tag Manager. Paid results stay separate from organic mentions and citations. [ChatGPT Ads →](skills/canonry/references/canonry-cli.md#openai-ads-chatgpt-ads) · [Google Ads →](docs/google-marketing.md)
  45. <a href="docs/google-marketing.md"><picture><source media="(max-width: 600px)" srcset="https://raw.githubusercontent.com/Canonry/canonry/main/docs/images/readme/ads-mobile.png" /><img src="https://raw.githubusercontent.com/Canonry/canonry/main/docs/images/readme/ads.png" alt="Google Ads performance: spend, conversions, cost per conversion, a daily chart, and campaign results, with sample data" width="100%" /></picture></a>
  46. <a id="compare-competitors-and-measure-complex-portfolios"></a>
  47. <a id="act-on-the-evidence"></a>
  48. **Also in the box:**
  49. - **[Aero](skills/aero/SKILL.md)** — A built-in analyst in the dashboard that reads your evidence and explains what changed.
  50. - **[Portfolios](skills/canonry/references/canonry-cli.md#target-measurement-plans)** — Versioned measurement plans for locations, products, or site sections. Branded and non-brand queries stay separate.
  51. - **[Competitor landscape](docs/competitor-landscape.md)** — Mention share, mentions, and citations for you and your competitors over time.
  52. - **[Query discovery and research](skills/canonry/references/canonry-cli.md#discovery-tracked-basket-expansion)** — Find new queries to track, or run one-off research without changing what you track.
  53. - **[Content engine](skills/canonry/references/canonry-cli.md#content-target-dismissals)** — Finds content gaps and the pages to write or improve next, so your agent can draft them and publish to WordPress for review.
  54. - **[Publish and index](skills/canonry/references/indexing.md)** — WordPress drafts, JSON-LD, and sitemap and URL submissions to Google and Bing.
  55. - **[Schedules and webhooks](skills/canonry/references/canonry-cli.md#scheduling--notifications)** — Rerun checks on a schedule and send alerts to Discord, Slack, or your systems.
  56. - **[Backlinks](skills/canonry/references/canonry-cli.md#backlinks-common-crawl)** — Common Crawl link data, queried locally with DuckDB.
  57. - **[Sentiment](docs/sentiment.md)** — Off by default. Rates branded answers as favorable, mixed, or unfavorable.
  58. - **And more** — browse the [documentation index](docs/README.md).
  59. **An example job for your connected agent:**
  60. > Find where competitors get cited and we don't. Use search data and the Agent Readiness audit scan to prepare a pull request or WordPress draft for review. After it ships, rerun checks and report what changed.
  61. ## How it works
  62. <p align="center">
  63. <a href="https://raw.githubusercontent.com/Canonry/canonry/main/docs/images/measure-act.svg">
  64. <img src="https://raw.githubusercontent.com/Canonry/canonry/main/docs/images/measure-act.svg" alt="AI, search, analytics, and traffic sources feed Canonry on schedules or on demand. Historical evidence across multiple domains and portfolios runs from baseline through the latest checks. Your agent reads evidence and sends commands to Canonry tools, while using its own tools for code and content. Canonry publishes to WordPress, submits to Google and Bing, and supplies the dashboard, evidence exports, and webhooks. Site changes loop back into measurement." width="100%" />
  65. </a>
  66. </p>
  67. ---
  68. ## Supported Engines
  69. Compare direct AI engines, OpenRouter models, and search answers from Exa and Parallel. Each keeps its own results and history.
  70. <p>
  71. <a href="https://platform.openai.com/api-keys"><kbd><img src="https://www.google.com/s2/favicons?domain=openai.com&amp;sz=64" alt="OpenAI logo" width="16" valign="middle" /> ChatGPT</kbd></a> &nbsp;
  72. <a href="https://console.anthropic.com/settings/keys"><kbd><img src="https://www.google.com/s2/favicons?domain=claude.ai&amp;sz=64" alt="Claude logo" width="16" valign="middle" /> Claude</kbd></a> &nbsp;
  73. <a href="https://aistudio.google.com/apikey"><kbd><img src="https://www.google.com/s2/favicons?domain=gemini.google.com&amp;sz=64" alt="Gemini logo" width="16" valign="middle" /> Gemini</kbd></a> &nbsp;
  74. <a href="https://www.perplexity.ai/settings/api"><kbd><img src="https://www.google.com/s2/favicons?domain=perplexity.ai&amp;sz=64" alt="Perplexity logo" width="16" valign="middle" /> Perplexity</kbd></a> &nbsp;
  75. <a href="https://dev.meta.ai/"><kbd><img src="https://www.google.com/s2/favicons?domain=meta.ai&amp;sz=64" alt="Meta logo" width="16" valign="middle" /> Meta (Muse)</kbd></a> &nbsp;
  76. <a href="docs/providers/openrouter.md"><kbd><img src="https://www.google.com/s2/favicons?domain=openrouter.ai&amp;sz=64" alt="OpenRouter logo" width="16" valign="middle" /> OpenRouter</kbd></a> &nbsp;
  77. <a href="docs/providers/answer-services/exa.md"><kbd><img src="https://www.google.com/s2/favicons?domain=exa.ai&amp;sz=64" alt="Exa logo" width="16" valign="middle" /> Exa</kbd></a> &nbsp;
  78. <a href="docs/providers/answer-services/parallel.md"><kbd><img src="https://www.google.com/s2/favicons?domain=parallel.ai&amp;sz=64" alt="Parallel logo" width="16" valign="middle" /> Parallel</kbd></a> &nbsp;
  79. <kbd>+ any OpenAI-compatible local model</kbd>
  80. </p>
  81. **Works with your agent:**
  82. <p>
  83. <a href="docs/plugins.md#install-for-claude-code"><kbd><img src="https://www.google.com/s2/favicons?domain=claude.ai&amp;sz=64" alt="Claude Code logo" width="16" valign="middle" /> Claude Code</kbd></a> &nbsp;
  84. <a href="docs/plugins.md#install-for-codex"><kbd><img src="https://www.google.com/s2/favicons?domain=openai.com&amp;sz=64" alt="Codex logo" width="16" valign="middle" /> Codex</kbd></a> &nbsp;
  85. <a href="docs/mcp.md"><kbd><img src="https://www.google.com/s2/favicons?domain=chatgpt.com&amp;sz=64" alt="ChatGPT logo" width="16" valign="middle" /> ChatGPT</kbd></a> &nbsp;
  86. <a href="docs/mcp.md"><kbd><img src="https://www.google.com/s2/favicons?domain=claude.ai&amp;sz=64" alt="Claude Desktop logo" width="16" valign="middle" /> Claude Desktop</kbd></a> &nbsp;
  87. <kbd>+ any MCP client or shell-capable agent</kbd>
  88. </p>
  89. <a id="built-in-integrations-and-workflows"></a>
  90. ## Integrations
  91. Connect the sources you use. Each one adds evidence or tools to the same project.
  92. <p>
  93. <a href="docs/google-search-console-setup.md"><kbd><img src="https://www.google.com/s2/favicons?domain=search.google.com&amp;sz=64" alt="Google Search Console logo" width="16" valign="middle" /> Search Console</kbd></a> &nbsp;
  94. <a href="docs/bing-webmaster-setup.md"><kbd><img src="https://www.google.com/s2/favicons?domain=bing.com&amp;sz=64" alt="Bing logo" width="16" valign="middle" /> Bing Webmaster Tools</kbd></a> &nbsp;
  95. <a href="docs/google-analytics-setup.md"><kbd><img src="https://www.google.com/s2/favicons?domain=analytics.google.com&amp;sz=64" alt="Google Analytics logo" width="16" valign="middle" /> Google Analytics 4</kbd></a> &nbsp;
  96. <a href="skills/canonry/references/google-business-profile.md"><kbd><img src="https://www.google.com/s2/favicons?domain=business.google.com&amp;sz=64" alt="Google Business Profile logo" width="16" valign="middle" /> Business Profile</kbd></a> &nbsp;
  97. <a href="docs/google-marketing.md"><kbd><img src="https://www.google.com/s2/favicons?domain=ads.google.com&amp;sz=64" alt="Google Ads logo" width="16" valign="middle" /> Google Ads</kbd></a> &nbsp;
  98. <a href="docs/google-marketing.md"><kbd><img src="https://www.google.com/s2/favicons?domain=tagmanager.google.com&amp;sz=64" alt="Google Tag Manager logo" width="16" valign="middle" /> Tag Manager</kbd></a> &nbsp;
  99. <a href="docs/cloudflare-traffic-setup.md"><kbd><img src="https://www.google.com/s2/favicons?domain=cloudflare.com&amp;sz=64" alt="Cloudflare logo" width="16" valign="middle" /> Cloudflare</kbd></a> &nbsp;
  100. <a href="skills/canonry/references/server-side-traffic.md"><kbd><img src="https://www.google.com/s2/favicons?domain=vercel.com&amp;sz=64" alt="Vercel logo" width="16" valign="middle" /> Vercel</kbd></a> &nbsp;
  101. <a href="skills/canonry/references/server-side-traffic.md"><kbd><img src="https://www.google.com/s2/favicons?domain=cloud.google.com&amp;sz=64" alt="Google Cloud logo" width="16" valign="middle" /> Cloud Run</kbd></a> &nbsp;
  102. <a href="docs/wordpress-setup.md"><kbd><img src="https://www.google.com/s2/favicons?domain=wordpress.org&amp;sz=64" alt="WordPress logo" width="16" valign="middle" /> WordPress</kbd></a> &nbsp;
  103. <a href="docs/seo-data.md"><kbd><img src="https://www.google.com/s2/favicons?domain=dataforseo.com&amp;sz=64" alt="DataForSEO logo" width="16" valign="middle" /> DataForSEO</kbd></a> &nbsp;
  104. <a href="skills/canonry/references/canonry-cli.md#backlinks-common-crawl"><kbd><img src="https://www.google.com/s2/favicons?domain=commoncrawl.org&amp;sz=64" alt="Common Crawl logo" width="16" valign="middle" /> Common Crawl</kbd></a> &nbsp;
  105. <a href="skills/canonry/references/canonry-cli.md#openai-ads-chatgpt-ads"><kbd><img src="https://www.google.com/s2/favicons?domain=chatgpt.com&amp;sz=64" alt="ChatGPT logo" width="16" valign="middle" /> ChatGPT Ads</kbd></a>
  106. </p>
  107. ---
  108. ## Quick start
  109. <a id="or-use-any-shell-capable-coding-agent"></a>
  110. ### With your agent
  111. Paste this into Claude Code, Codex, or any shell-capable coding agent.
  112. <details>
  113. <summary>Copy the first-time setup request</summary>
  114. ```text
  115. Help me set up Canonry for my public site.
  116. Use the official Canonry docs:
  117. - Agent quickstart: https://github.com/Canonry/canonry#or-use-any-shell-capable-coding-agent
  118. - CLI reference: https://github.com/Canonry/canonry/blob/main/skills/canonry/references/canonry-cli.md
  119. - Plugin setup: https://github.com/Canonry/canonry/blob/main/docs/plugins.md
  120. - MCP setup: https://github.com/Canonry/canonry/blob/main/docs/mcp.md
  121. Pick one path and stay on it. Use connected Canonry tools (plugin or MCP) only if you have them and I have not pointed you at a specific install. A connected tool runs in its own process and never sees `CANONRY_CONFIG_DIR`, `CANONRY_PORT`, or anything else you export in a shell, so it always acts on my default install. If I named a config directory, a port, or a sandbox, that is the shell path: use `cnry` and tell me which path you chose. Never mix the two in one run, and never create a duplicate project. The `cnry` and `canonry` commands are interchangeable.
  122. 1. Ask for my public domain, country, and language. Do not create or scan anything yet. Country and language are only applied when a project is created, so if you reuse an existing project, report its country and language instead of changing them.
  123. 2. Shell path only: confirm `cnry` is on PATH, then run `cnry --version`. If Canonry is missing, propose `npm install -g @canonry/canonry` and wait for approval. Then run `cnry doctor --format json`, which is the command that says whether config, database, and server are in place. `cnry --version` does not read config and succeeds on a completely unconfigured install, so it cannot answer this. If config is missing, run `cnry bootstrap` yourself: it is not interactive, takes about a second, and is safe to rerun. Do not hand it to me and wait. The interactive command is `cnry init`, which is optional provider and OAuth setup that Agent Readiness does not need. Bootstrap prints an API key, so do not repeat its output back to me, and never ask me to paste passwords, API keys, OAuth credentials, or command output.
  124. 3. Confirm the API is reachable. `cnry doctor --format json` reports it, and any project read exits non-zero with `CONNECTION_ERROR` when it is not. If it is unreachable, propose `cnry start` and wait for approval. Use `cnry start`, which is the background daemon, and not `cnry serve`, which runs in the foreground and will block you until I stop it, even though some error messages suggest it. Stop anything you started with `cnry stop`.
  125. 4. List projects with the connected project tool or `cnry project list --format json`, and reuse one whose domain matches. Confirm the proposed name is not already assigned to a different domain. To find out whether a project has already been scanned, run `cnry agent-readiness score <project> --format json` with no `--run-id`, which reports the latest run. Read the `hasData` field, not the score: this command exits 0 and reports `aggregateScore: 0` for a project that has never been scanned, so reading the score alone would have you tell me my site scored zero. If `hasData` is true and `runStatus` is `completed` or `partial`, read that scan instead of starting a new one. If no project matches, show the exact create operation and wait for approval.
  126. 5. Propose a bounded Agent Readiness audit scan: `--max-pages 100` for a first look, plus whether dead-link checking is on (it is off unless you pass `--check-dead-links`). Show the connected operation or the exact `cnry agent-readiness run <project> --max-pages 100 --wait --format json` command with the project name filled in, and wait for separate approval before scanning. `--wait` polls for up to 15 minutes and returns only the run id and status. If the status it returns is still `queued` or `running`, the scan has not finished: do not rerun it or report results, poll `cnry agent-readiness progress <project> --run-id <run-id> --format json` until it is terminal. If `--wait` outruns your own tool timeout first, recover the run id with `cnry runs <project> --kind site-audit --limit 1 --format json`, which returns the newest scan whatever its status, and poll the same way.
  127. 6. When the run is `completed` or `partial`, read `cnry agent-readiness overview <project> --run-id <run-id> --format json` first. It is the only one of these commands that carries `termination` and `complete`; the score and page-scores commands do not. Then read `cnry agent-readiness score <project> --run-id <run-id> --format json` and `cnry agent-readiness page-scores <project> --run-id <run-id> --sort score-asc --limit 10 --format jsonl`. Tell me the termination reason in plain words and whether the scan covered the whole site or stopped at a page, link, depth, or time limit. A `partial` run scored the pages it reached and not my site, so never present it as a full-site result. If it stopped early, the fix depends on the reason, so say which: a page or depth limit needs a larger budget, a time limit needs a smaller scan (a lower `--max-pages` or `--max-depth`). If the run failed or was cancelled, inspect the run error and stop.
  128. 7. Summarize only completed evidence, then propose AI Visibility setup. Ask before you add queries, connect providers, start a provider-backed or quota-consuming run, edit files, or publish.
  129. ```
  130. </details>
  131. ### With the plugin
  132. ```bash
  133. # Claude Code
  134. claude plugin marketplace add Canonry/canonry
  135. claude plugin install canonry@canonry
  136. # Codex
  137. codex plugin marketplace add Canonry/canonry
  138. codex plugin add canonry@canonry
  139. ```
  140. Or connect the [MCP adapter](docs/mcp.md) to any MCP client. Full steps: [Agent Plugin guide](docs/plugins.md).
  141. <a id="get-a-page-health-baseline"></a>
  142. ### On your own
  143. ```bash
  144. npm install -g @canonry/canonry
  145. cnry bootstrap
  146. cnry serve
  147. ```
  148. Then open [http://127.0.0.1:4100/setup](http://127.0.0.1:4100/setup) and follow the setup to scan your site. The crawl saves an Agent Readiness baseline. AI Visibility is optional and has a separate setup.
  149. <details>
  150. <summary>Scan and read results from the terminal</summary>
  151. Keep `cnry serve` active. In a second terminal, create a project and start a bounded scan:
  152. ```bash
  153. cnry project create my-site --domain example.com --country US --language en
  154. cnry agent-readiness run my-site --max-pages 100 --wait --format json
  155. ```
  156. Read the run ID and status from the output. If the status is `completed` or `partial`, read evidence from that run:
  157. ```bash
  158. cnry agent-readiness score my-site --run-id <run-id> --format json
  159. cnry agent-readiness page-scores my-site --run-id <run-id> --sort score-asc --limit 10 --format jsonl
  160. ```
  161. `--wait` polls for up to 15 minutes. If the scan remains active, use `cnry agent-readiness progress <project> --run-id <id> --format json`.
  162. If the scan fails or is cancelled, read the error with `cnry run show <run-id> --format json`.
  163. </details>
  164. <details>
  165. <summary>Configure a provider and run visibility checks</summary>
  166. For an existing project, configure a provider with your own key.
  167. | Provider | Key source | Environment variable |
  168. |---|---|---|
  169. | Gemini | [Google AI Studio](https://aistudio.google.com/apikey) | `GEMINI_API_KEY` |
  170. | OpenAI | [OpenAI Platform](https://platform.openai.com/api-keys) | `OPENAI_API_KEY` |
  171. | Meta (Muse) | [Meta Model API](https://dev.meta.ai/) | `MUSE_API_KEY` |
  172. | Claude | [Anthropic Console](https://console.anthropic.com/settings/keys) | `ANTHROPIC_API_KEY` |
  173. | Perplexity | [Perplexity settings](https://www.perplexity.ai/settings/api) | `PERPLEXITY_API_KEY` |
  174. | [OpenRouter](docs/providers/openrouter.md) | [OpenRouter keys](https://openrouter.ai/keys) | `OPENROUTER_API_KEY` |
  175. | [Exa](docs/providers/answer-services/exa.md) | [Exa dashboard](https://dashboard.exa.ai/api-keys) | `EXA_API_KEY` |
  176. | [Parallel](docs/providers/answer-services/parallel.md) | [Parallel platform](https://platform.parallel.ai) | `PARALLEL_API_KEY` |
  177. | Local model | Any OpenAI-compatible endpoint | `LOCAL_BASE_URL` |
  178. Add the queries that matter to your project:
  179. ```bash
  180. cnry query add my-site "your first query" "your second query"
  181. cnry run my-site --wait
  182. cnry visibility-stats my-site --by-provider
  183. ```
  184. </details>
  185. <a id="why-we-use-model-providers-apis-directly-not-a-router"></a>
  186. <details>
  187. <summary>Direct APIs, OpenRouter, and search answer services</summary>
  188. Canonry supports direct model APIs, OpenRouter, and search answer services such
  189. as Exa and Parallel. Direct adapters retain the search and grounding details
  190. each provider exposes. OpenRouter offers multiple models through one key. Exa
  191. and Parallel search the web and return answers with source evidence.
  192. Each direct engine, OpenRouter model route, and service profile keeps separate
  193. results and history. The same model through different APIs can return different
  194. answers and citation evidence. Canonry measures brand mentions separately from
  195. source citations and keeps missing evidence unknown.
  196. [Provider guides](docs/providers/README.md) describe the evidence and limits of
  197. each integration.
  198. </details>
  199. <details>
  200. <summary>Discover questions or run one-off research</summary>
  201. **Expand your tracked questions.** Run `cnry discover run <project> --icp "..."`. This does not change the tracked questions.
  202. Preview a completed session with `cnry discover promote preview <project> <session-id>`. Promote only after approval.
  203. **Research without changing tracking.** Run `cnry research run <project> "query one" "query two" --wait`.
  204. For a configured portfolio destination, select one `--market <key>` or `--property <key>`; Canonry saves the destination with the answers without changing the query. Add paired `--template-id <id> --template-version <version>` only to preserve an already-expanded template's provenance. Research never adds queries to the tracked questions.
  205. </details>
  206. <a id="if-you-get-stuck"></a>
  207. <details>
  208. <summary>Troubleshoot installation, scans, and visibility results</summary>
  209. | Problem | Next step |
  210. |---|---|
  211. | Agent Readiness audit scan is still active | Read exact counters with `cnry agent-readiness progress <project> --run-id <id> --format json`. |
  212. | Agent Readiness audit scan failed | Read the error with `cnry run show <run-id> --format json`. Read the last phase and counters with the progress command. |
  213. | No visibility results | Inspect existing work with `cnry runs <project> --format json`, then `cnry run show <run-id> --format json`. This does not start another paid run. |
  214. | `npm install` fails on `node-gyp` | Install build tools for `better-sqlite3` ([guide](https://github.com/WiseLibs/better-sqlite3/blob/master/docs/troubleshooting.md)). |
  215. </details>
  216. <a id="deployment-and-trust-boundary"></a>
  217. <a id="technical-surface"></a>
  218. ## Self-hosting and API
  219. Canonry is single-tenant. Run one instance for one operator or team. Keep unrelated teams on separate instances.
  220. The CLI and REST API are the primary interfaces. They expose measurements, diagnoses, actions, evidence exports, and schedules.
  221. OpenAPI is available at `GET /api/v1/openapi.json`.
  222. See the [deployment guide](docs/deployment.md) for reverse proxies, daemon mode, Docker, systemd, and Tailscale.
  223. ## Documentation
  224. - **New to Canonry:** [Getting started: the first-run loop and questions to ask](skills/aero/references/getting-started.md)
  225. - **Operate Canonry:** [CLI reference](skills/canonry/references/canonry-cli.md) · [Agent Plugin](docs/plugins.md) · [MCP](docs/mcp.md)
  226. - **Use built-in Aero:** [Setup and analysis workflows](skills/aero/SKILL.md)
  227. - **Understand the system:** [Architecture](docs/architecture.md) · [Data model](docs/data-model.md)
  228. - **Install standalone agent skills:** `cnry skills install` ([guide](skills/canonry/SKILL.md))
  229. - **Browse all guides:** [Documentation index](docs/README.md)
  230. ---
  231. <a id="support"></a>
  232. ## Community &amp; Support
  233. - **Discord:** Join the community on **[Discord](https://discord.gg/8K9yY7GywJ)** for help and questions.
  234. - **Twitter / X:** Follow **[@canonry_ai](https://x.com/canonry_ai)** for updates.
  235. - **Feedback &amp; Ideas:** Missing something? [Open an issue](https://github.com/Canonry/canonry/issues).
  236. - **Show Support:** [Star](https://github.com/Canonry/canonry) this repo to follow along.
  237. ---
  238. ## Contributing
  239. ```bash
  240. git clone https://github.com/Canonry/canonry.git && cd canonry
  241. pnpm install
  242. ```
  243. For local agent work, run tests only after changes are finalized and ready for push or review.
  244. `pnpm verify` is optional locally. See [CONTRIBUTING.md](./CONTRIBUTING.md) and the [testing guide](docs/testing.md).
  245. <a href="https://github.com/Canonry/canonry/graphs/contributors">
  246. <img src="https://contrib.rocks/image?repo=Canonry/canonry" alt="Canonry contributors" />
  247. </a>
  248. ## License
  249. [MIT](./LICENSE), starting with Canonry 5.0.0. Use, modify, self-host, redistribute, and build commercial products with Canonry. Keep the copyright and permission notice with copies or substantial portions of the software.
  250. Earlier releases retain their original licenses, including any future-license grants. Dependencies and bundled assets retain their own licenses; see [third-party notices](packages/canonry/THIRD_PARTY_NOTICES.md).

Ratings and comments

Sign in to rate this skill.

Sign in

Free to use. If it’s helped, pass it on.