{"slug": "find-answers-faster-with-pagerduty-docs-one-site-for-engineers-and-the-ai-they", "title": "Find answers faster with PagerDuty Docs: one site for engineers and the AI assistants they work with by Aatharsha Jeyachelvan", "summary": "PagerDuty launched the new PagerDuty Docs at docs.pagerduty.com on October 7, 2026, consolidating documentation that had been spread across 21 subdomains and four publishing platforms into a single site with one search and 1,200 pages. The site adds a machine-readable layer with a Markdown counterpart for every page plus llms.txt and llms-full.txt so AI assistants read the documentation directly, alongside one-click actions to send any page into an AI assistant. PagerDuty cited that 68% of developers turn to docs first when learning a new tool and 84% use an AI tool daily as the reason documentation must now serve both engineers and machines.", "body_md": "- [PagerDuty](https://www.pagerduty.com/) /\n- [Blog](https://www.pagerduty.com/blog) /\n- [Announcements](https://www.pagerduty.com/blog/category/announcements/) /\n- Find answers faster with PagerDuty Docs: one site for engineers and the AI assistants they work with\n\n# Blog\n\n# Find answers faster with PagerDuty Docs: one site for engineers and the AI assistants they work with\n\n[Aatharsha Jeyachelvan](https://www.pagerduty.com/blog/author/ajeyachelvan/)October 7, 2026 | 6 min read\n\n*This blog post is part of PagerDuty’s ongoing series on how we’re helping customers navigate their journey towards autonomous operations. Read on to learn about how PagerDuty’s PagerDuty Docs recently announced in [GA] builds towards this vision.*\n\nDocumentation is the living manual for any product. Over 68% of developers still turn to docs first when learning a new tool, and 84% now use an AI tool daily. Those two numbers together change what documentation is for. Increasingly, the assistant reads the manual and the engineer reads the assistant. If the manual is hard to find, split across sites, or wrapped in layout the model has to fight through, the answer that comes back is a guess.\n\nUntil now, PagerDuty’s documentation lived across 21 subdomains and four publishing platforms: the Support KB in one place, developer docs in another, integration guides and Ops Guides somewhere else, each with its own brand treatment, search box, and navigation. There was no shared search, no shared information architecture, and no single canonical source. If our own teams struggled to find the right page, a customer on call at 2 a.m. stood even less of a chance, and the next stop after a wrong guess is a support ticket. The shift underway across the industry is to treat documentation as infrastructure with two audiences, people and machines, and to design for both at once rather than bolting a chatbot onto a site that was never structured to be read by one.\n\nToday we’re introducing the new PagerDuty Docs at docs.pagerduty.com: one site, one search, one source of truth. All 1,200 pages now live in a single architecture with instant search across the full corpus, a consistent page layout with one-click actions to take any page into your AI assistant, and a machine-readable layer underneath, with a Markdown counterpart for every page, llms.txt, and llms-full.txt, so assistants read our documentation instead of reconstructing it. Let’s look at each.\n\n**Navigate by what you’re trying to do, not where the answer happens to live**\n\nThe old sites were split along publishing-platform lines. That meant you had to know in advance whether your answer was a “support” question or a “developer” question before you could even pick a search box. Most real tasks are both: you read the concept, then you hit the API.\n\nThe new homepage is organized around the three pillars of the Operations Cloud: Full Lifecycle Incident Management, AI & Automation, and Platform and Ecosystem. Under each pillar the entry points are action-based and follow your journey with PagerDuty: set something up, integrate something, automate something, troubleshoot something. Pick a pillar and the navigation narrows with you. Support content and developer content sit side by side in the same tree, so a page on shift-based schedules is one click from the corresponding API reference rather than a separate site away.\n\n**Read a page, take it with you, and search from anywhere**\n\nFour publishing platforms meant four brand treatments, four code-block styles, and four different ideas of where the table of contents goes. Every context switch cost a few seconds of reorientation, and people live in these pages for hours.\n\nEvery page now uses the same layout regardless of where it originally came from. The left rail shows where you are in the tree, the right rail is the on-page table of contents so you can jump straight to the section you need, and the middle is deliberately plain content. Above the content sit the page actions:\n\n- Copy as Markdown: grab the page in a format a model can parse cleanly and paste it into whatever tool you’re working in.\n- View raw: see the source content with no chrome.\n- Open in your assistant of choice: hand the page directly to an AI assistant in one click.\n\nLight and dark mode are both supported and the choice persists across visits. Search is available from every page via ⌘K or the search field. It’s instant and runs client-side, and it covers the full 1,200-page corpus, so you never have to return to a homepage to look something up.\n\n**Give AI assistants the manual, not the layout**\n\nThis is the part that addresses the opening problem. When an assistant fetches a typical docs page it gets navigation, header chrome, scripts, and cookie banners wrapped around the content it actually wants. When it can’t reach the docs at all, it falls back to whatever it picked up elsewhere, and the person on the other end gets a confident answer that’s out of date or wrong.\n\nPagerDuty Docs ships a machine-readable layer alongside the human one:\n\n- A Markdown counterpart for every page. Append .md to any page URL and you get the same content with navigation, chrome, and JavaScript stripped out. An assistant fetching this gets the answer, not our layout.\n- /llms.txt: a site map written for machines. Every section and every page, each with a description, so an assistant landing there knows exactly what we have and where it is.\n- /llms-full.txt: the entire documentation set as a single plain-text file. One fetch, the whole manual, for tools that want to load everything into context at once.\n- Open crawling and structured data. robots.txt is open, structured data is on every page, and nothing sits behind a crawler block.\n\nThe result is that when someone asks their assistant a PagerDuty question, the assistant is reading our documentation accurately and in full, instead of guessing.\n\n**Bringing it together**\n\nEach piece on its own is a modest improvement. Together they close the loop between the two ways engineers now consume documentation. The unified architecture and single search mean a person can find the right page without knowing which of 21 subdomains it used to live on. The page actions let that person carry the page into their assistant in a format the model can parse. And the Markdown counterparts, llms.txt, and llms-full.txt mean the assistant can go get the page itself, with the same canonical content the human would see.\n\nThat’s the Platform and Ecosystem work in a nutshell: fewer places where a wrong guess turns into an incident or a support ticket, and a documentation surface that the growing set of AI tools in your workflow can rely on. The engineer stays in the loop; the assistant just has better material to work from.\n\nPagerDuty’s approach to autonomous operations:\n\n- Puts intelligent agents to work at scale — handling the noise, accelerating resolution, and keeping you in the loop where it matters\n- Deepens the full incident management lifecycle by empowering teams to resolve incidents faster\n- Broadens the platform and ecosystem with capabilities that help teams prevent incidents from happening\n\nPagerDuty Docs moves us closer to that vision. Ready to see how it can help your team find answers before they become incidents? Watch the demo!\n\nExplore the new site at docs.pagerduty.com. Learn more at pagerduty.com/platform or connect with your PagerDuty account team.", "url": "https://wpnews.pro/news/find-answers-faster-with-pagerduty-docs-one-site-for-engineers-and-the-ai-they", "canonical_source": "https://www.pagerduty.com/blog/announcements/find-answers-faster-with-pagerduty-docs-one-site-for-engineers-and-the-ai-assistants-they-work-with/", "published_at": "2026-10-07 11:30:57+00:00", "updated_at": "2026-10-07 12:18:54.425942+00:00", "lang": "en", "topics": ["ai-crawlers", "ai-search", "structured-data", "ai-products"], "entities": ["PagerDuty", "PagerDuty Docs", "docs.pagerduty.com", "Aatharsha Jeyachelvan", "Operations Cloud"], "also_reported_by": [], "alternates": {"html": "https://wpnews.pro/news/find-answers-faster-with-pagerduty-docs-one-site-for-engineers-and-the-ai-they", "markdown": "https://wpnews.pro/news/find-answers-faster-with-pagerduty-docs-one-site-for-engineers-and-the-ai-they.md", "text": "https://wpnews.pro/news/find-answers-faster-with-pagerduty-docs-one-site-for-engineers-and-the-ai-they.txt", "jsonld": "https://wpnews.pro/news/find-answers-faster-with-pagerduty-docs-one-site-for-engineers-and-the-ai-they.jsonld"}}