Agent View
The site as an AI agent gets it, beside the site as a person sees it. What it can read, what it has to cut away, and what it never sees.
Sign in with MuckRock
Upload.
Analyze.
Embed.
DocumentCloud is an all-in-one platform used by newsrooms around the world for work with primary source documents.
Upload and organize documents.
Extract, analyze and search text.
Annotate, redact and edit.
Embed documents on your website.
7,265,633 public documents and counting
Everyone is welcome to explore our public document archive and organize interesting documents into projects.
Trending documents
- Carta sobre as eleições presidenciais de 2026 2º turno V5 (1)
The rest of the page is in full screen.
[Sign in with MuckRock](https://api.www.documentcloud.org/accounts/login/squarelet?next=https://www.documentcloud.org/) # Upload. Analyze. Embed. **DocumentCloud** is an all-in-one platform used by newsrooms around the world for work with primary source documents. Upload and organize documents. Extract, analyze and search text. Annotate, redact and edit. Embed documents on your website. ## 7,265,633 public documents and counting Everyone is welcome to explore our public document archive and organize interesting documents into projects. ### Trending documents - [Carta sobre as eleições presidenciais de 2026 2º turno V5 (1)](https://www.documentcloud.org/documents/28829521-carta-sobre-as-eleicoes-presidenciais-de-2026-2o-turno-v5-1/) - [District Communication 10.19.26](https://www.documentcloud.org/documents/28824159-district-communication-101926/) - [MAP - 10-17-26](https://www.documentcloud.org/documents/28821199-map-10-17-26/) - [Letter - Sergeant-at-Arms - Dated 10-22-26](https://www.documentcloud.org/documents/28829317-letter-sergeant-at-arms-dated-10-22-26/) - [Louise Lucas federal indictment](https://www.documentcloud.org/documents/28828964-louise-lucas-federal-indictment/)  ## Documents from trustworthy sources Newsrooms, independent journalists, archives and academic projects can upload, annotate, and publish documents. To get started, create a MuckRock account and request verification. [Create an account](https://accounts.muckrock.com/accounts/signup/?intent=documentcloud&next=https://www.documentcloud.org/home/) [Learn more about verification](https://help.muckrock.com/Request-verification-19ef8892696381dba944e17e14938433) ## Turn documents into data ### Public API DocumentCloud has a [public API](https://www.documentcloud.org/help/api/) that you can use to manage large-volume projects. ### Add-ons Use community-authored extensions and tools help you do even more with your documents and data. ### Full-text search Automatic OCR and search indexing make it easy to explore documents. ### Python SDK Get started with our [Python SDK](https://documentcloud.readthedocs.io/en/latest/) to automate document uploads, analysis, and processing tasks. DocumentCloud is an open-source project. [Explore the code on GitHub.](https://github.com/MuckRock/documentcloud) DocumentCloud is a project of the [MuckRock Foundation](https://www.muckrock.com/), a 501c3 organization that is committed to trust, transparency, and civic engagement. Our [privacy policy](https://www.muckrock.com/privacy-policy/) and [terms of service](https://www.muckrock.com/tos/) apply to all users on this site. - [About MuckRock](https://www.muckrock.com/) - [Job Openings](https://www.muckrock.com/jobs/) ## Together, we hold power to account and shine a light on government transparency ### Your donation fuels vital tools such as MuckRock Requests, FOIA Machine, DocumentCloud and oTranscribe that promote civic engagement. - To donate via Apple Pay, [visit our dedicated GiveButter page](https://www.givebutter.org/muckrock/). - Or mailed to: MuckRock Foundation 263 Huntington Ave Boston, MA 02115 You can always find our most current financial statements at [MuckRock's financial transparency page](https://www.muckrock.com/financial/). If you'd like to support our work with a legacy, a gift of stocks or crypto currency, contact [[email protected]](mailto:[email protected]).
No <main>: Wrap the page’s own content in <main>, so a reader knows where it starts and stops, and Clean can be measured.
The figures
8 pages, set against the 167,985 sites with figures.
- Readablehigher is better
- 88%7 of 8 pages
- All 167,985 sitesBetter than 28%
Pages whose own text is in the HTML the server sends, enough of it for their kind.
- Cleanhigher is better
- 26%middle page
- All 167,985 sitesBetter than 13%
Of the words before script, the share that are the page and not menus, footer and forms. 2 pages are not measured, with no <main> or no words.
- Costshown, never scored
- 1,723tokens
- About 215 a page
To read every page once. A long page that answers well is not worse for it. Tokens are estimated at about 4 characters each.
How its pages spread on Clean
6 pages measured. Not measured: 1 with no <main> or no words and 1 shell with no words before script.
Against its rivals
The sites nearest it by what they do, with figures. Close means similar work, not better.
| Readinghigher is better | Costnot scored | ||
|---|---|---|---|
| Site | Readable | Clean | Tokens a page |
| documentcloud.orgThis site, 8 pages | 88% | 26% | 215 |
| centermarkconstruction.com46 pages | 76% | 52% | 295 |
| free.tools83 pages | 100% | 95% | 899 |
| theicstock.com100 pages | 100% | 2% | 2,085 |
Third of 4 on Readable, third on Clean.
AI crawler access
What robots.txt says to each, as read on October 24, 2026. 3 of 8 may read the site.
| Crawler | robots.txt says | Used for | The line that decides |
|---|---|---|---|
| GPTBotOpenAI | Blocked | Training models | User-agent: GPTBot / Allow: /home/$ |
| OAI-SearchBotOpenAI | Not named | ChatGPT search | Falls to User-agent: * |
| ChatGPT-UserOpenAI | Blocked | A page a person asks about | User-agent: ChatGPT-User / Allow: /home/$ |
| ClaudeBotAnthropic | Blocked | Training models | User-agent: ClaudeBot / Allow: /home/$ |
| Claude-SearchBotAnthropic | Not named | Claude’s search | Falls to User-agent: * |
| PerplexityBotPerplexity | Blocked | Perplexity search | User-agent: PerplexityBot / Allow: /home/$ |
| Google-ExtendedGoogle | Blocked | Training Gemini | User-agent: Google-Extended / Allow: /home/$ |
| bingbotMicrosoft | Not named | Bing and Copilot | Falls to User-agent: * |
5 of 8 are kept out. A crawler kept out never reads a page, however well it is made.
Every page
Worst first. Each opens in Agent View at the top.
3 pages of 8 lose something. Pick a loss to see what to do about it.Open the action plan
| What an agent getshigher is better | Signals | Costnot scored | ||||
|---|---|---|---|---|---|---|
| Page | Words | Clean | Readable | Main heading | Tokens | What it loses |
| /API-19ef889269638147bbb7d8cc8af8e0fc | 0 | No words | No | None | 0 | Shell |
| /home | 327 | No <main> | Yes | Yes | 1,035 | No <main> |
| /documents | 1 | 4% | Yes | None | 2 | No h1 |
| /documents/28829521-carta-sobre-as-eleicoes-presidenciais-de-2026-2o-turno-v5-1 | 10 | 23% | Yes | Yes | 169 | Nothing |
| /documents/28829317-letter-sergeant-at-arms-dated-10-22-26 | 10 | 26% | Yes | Yes | 142 | Nothing |
| /documents/28828964-louise-lucas-federal-indictment | 10 | 28% | Yes | Yes | 134 | Nothing |
| /documents/28821199-map-10-17-26 | 10 | 29% | Yes | Yes | 110 | Nothing |
| /documents/28824159-district-communication-101926 | 10 | 29% | Yes | Yes | 131 | Nothing |
Show where it stands
A card, a badge and a line, each with the figures.
A badge for the site
<a href="https://mttr.si/agent-view/documentcloud.org"><img src="https://mttr.si/agent-view/documentcloud.org/badge.svg" alt="Agent View: 88% readable, 26% clean" height="20"></a>
The figures in one line
documentcloud.org in Agent View: 88% of pages readable, better than 28% of sites; 26% clean; 1,723 tokens to read all 8 pages. mttr.si/agent-view/documentcloud.org
What to do next
Fix the pages, let an agent do it, and read the site again to see the figures move.
Fix what agents miss
- Put the text of 1 page in the HTMLReadable
- Give 1 page a main headingClearer
- Wrap 1 page in <main>Clean measured
Connect your agent
Claude Code, Codex or Cursor reads the site’s fixes over MCP, fixes a page in your code, has mttr read it again, and checks the fix landed.
https://mttr.si/mcpSet up an agent
Claim the site
Prove it is yours with a DNS record or a file. Your agent can then reach its reads and jobs, and set each change to wait for your approval.
Claim the siteDrawn from the HTML mttr fetched on its read of October 24, 2026, not a screenshot, and only from the 8 pages read. An assistant’s own reader may differ. Tokens are estimated at about 4 characters each. A site an agent reads well is not a site an assistant cites.