# GraphStaff: full website context > GraphStaff is an AI knowledge platform for developer documentation, customer knowledge hubs, and a permission-aware Company Brain. It gives people and AI agents grounded answers from one governed knowledge graph. Generated from the canonical marketing pages on 2026-07-29. ## Organization facts - Name: GraphStaff - Canonical website: https://graphstaff.ai/ - Product category: AI knowledge platform - Core products: Developer Docs and API Docs, Customer Knowledge Hubs, and Company Brain - Core differentiator: one governed knowledge graph supplies grounded, permission-aware answers to people and AI agents - Contact: use the Request a Demo form on the canonical website ## Canonical page content ## AI Knowledge Platform | GraphStaff Canonical URL: https://graphstaff.ai/ Summary: GraphStaff is an AI knowledge platform for developer docs, customer knowledge hubs, and company knowledge, with governed answers for people and AI agents. ### Knowledge, readyfor anyone who asks. Customers, developers, employees, and their AI agents all ask. GraphStaff gives them one trusted answer. The direct answer #### What is GraphStaff? GraphStaff is an AI knowledge platform that turns developer documentation, customer help content, and internal company knowledge into one governed graph. People and AI agents ask questions through websites, search, Ask AI, or MCP, and GraphStaff returns grounded answers while applying the same permissions across every surface. Everyone asks #### All day, from every direction, someone is asking. Different askers, same need: an answer they can trust. A customer · Help center “Can I download invoices as a PDF?” A coding agent · MCP “How do I paginate through all customers?” An AI assistant · Company Brain “What changed in the March release?” A prospect · Website “Do you support SAML SSO with Okta?” A developer · API reference “What causes a 429 rate limit error?” A teammate · Slack “How do I file an expense over $500?” A support agent · Console “Which plans include audit logs?” An on-call engineer · Runbook “How do I open a production incident?” SUPPORT MACRO · UPDATED LAST YEAR Export your data under Settings, then Data, then Export. docs/exports.mdx · 14 MONTHS AGO Data exports run from the account dashboard. CHAT WIDGET Sure! You can export straight from the mobile app. unverified ANA · #support which one of these is actually current? CODING AGENT GET /v2/exports returns 404. The reference still documents /v1/exports. #### They all ask the same questions. Today they get different answers. #### GraphStaff connects your knowledge into one governed graph. Every answer, for every asker, comes from it. PEOPLE Customers · developers · teams Governed knowledge One graph · one policy AI AGENTS Claude · ChatGPT · your stack SIX SURFACES · ONE POLICY Website Search Ask AI MCP llms.txt Markdown #### Start with the audience that matters now. [01 For your developers A developer · and their coding agent “How do I verify webhook signatures?” ##### Developer Docs and API Docs Guides, API reference, and grounded answers, with MCP access for the coding agents they already use. Explore the product](https://graphstaff.ai/developer-docs/) [02 For your customers A customer · Help center “How do I add a teammate to my account?” ##### Knowledge Hubs A branded help center where customers search, browse, or ask, and find the answer before they open a ticket. Explore the product](https://graphstaff.ai/knowledge-hubs/) [03 For your team and its AI A teammate · asking Claude “How many PTO days carry over?” ##### Company Brain The same governed company context for Claude, ChatGPT, and every approved internal AI. Explore the product](https://graphstaff.ai/company-brain/) Similarity search One passage that sounds right Graph retrieval The neighborhood the answer depends on Our Secret Sauce #### Similarity finds a passage. Relationships build an answer. The graph brings in what an answer depends on: definitions, prerequisites, policies, and exceptions. [See how the engine works](https://graphstaff.ai/our-secret-sauce/) AM Alex Morgan Product · customers group GraphStaff policy Allowed context VISIBLE TO ALEX AND THEIR AI AGENT Product roadmap · Customer guidance NOT RETRIEVED Finance runbook · Executive notes CONTROL WHO KNOWS WHAT #### Everyone asks the same brain. No one sees what they should not. Identity and group rules ship on every plan, and one policy governs the website, search, Ask AI, and MCP retrieval. [How permission-aware AI works](https://graphstaff.ai/company-brain/) Questions, answered #### Frequently asked questions. The short version of what teams usually want to know before they see GraphStaff in action. #### How is GraphStaff different from a traditional knowledge base? A traditional knowledge base mainly stores and publishes pages. GraphStaff also connects pages, concepts, and policies into a governed graph, then serves grounded answers through websites, search, Ask AI, MCP, llms.txt, and Markdown. #### Can we start with only one product? Yes. Each product stands on its own. A team can begin with developer docs, a customer knowledge hub, or a Company Brain and expand only when another use case becomes useful. #### Does GraphStaff work with AI tools we already use? Yes. GraphStaff provides an MCP server and other machine-readable surfaces so your knowledge can work with Claude, ChatGPT, coding agents, and your own AI infrastructure without committing to one model vendor. #### How do we see whether it fits? Request a demonstration and tell us which audience and knowledge set matter most. We can plan a focused walkthrough or a proof of concept around your actual use case. Your knowledge is ready #### Give it somewhereuseful to go. Start with the audience that matters now. GraphStaff will make the same knowledge ready for everything that comes next. --- ## Developer Docs and API Docs | GraphStaff Canonical URL: https://graphstaff.ai/developer-docs/ Summary: Developer documentation and API references for humans and AI agents, with grounded Ask AI, automatic MCP, llms.txt, and Markdown pages. Developer Docs and API Docs ### Documentation built for builders and the AI agents building with them. One polished home for guides, API reference, and grounded answers, with native access for the coding agents your developers already use. The direct answer #### How does GraphStaff make documentation usable by AI agents? GraphStaff publishes the same maintained documentation through an MCP server, llms.txt and llms-full.txt indexes, per-page Markdown, search, and grounded Ask AI. Coding agents retrieve current guides and API references directly, while identity and group permissions keep private pages out of unauthorized answers. Documentation for people #### A reading experience that earns trust. Guides, API reference, examples, and search stay in one coherent interface, and every answer stays connected to a page your team owns. docs.yourcompany.com Live YOUR DOCS One maintained source Guides · APIs · examples MCP server llms.txt Markdown pages Website Ask AI AI TOOLS Native documentation access Search · list · grep · fetch Documentation for AI agents #### Readers get a website. AI agents get MCP. GraphStaff publishes MCP, llms.txt, and Markdown interfaces beside the human site, so AI agents retrieve exact, current pages instead of scraping and guessing. An MCP server to hand out #### Your docs get their own MCP server. Give it to your customers. GraphStaff spins it up automatically when you publish. Customers add it to Claude, ChatGPT, or their coding agent, and the AI agent reads your current reference and writes working code for your product. Your customer’s coding agent docs.yourcompany.com/mcp Add plan upgrades to our billing page. mcp · search("update subscription plan") → Update a subscription · API reference Your docs route plan changes through the subscriptions endpoint: await yourco.subscriptions.update(id, { plan: "pro" }) Written from your current reference Private docs included #### Private docs ship on every plan. Protect partner references, beta docs, and internal knowledge. The same permissions follow the reader into search, Ask AI, and AI agent retrieval. docs.yourcompany.com Visitor Partner Quickstart guide Public ✓ ✓ Core API reference Public ✓ ✓ Partner API endpoints Partners only – ✓ Beta feature docs Beta program – ✓ Same rules in Search Ask AI MCP Questions, answered #### Frequently asked questions. The short version of what teams usually want to know before they see GraphStaff in action. #### Does GraphStaff support both guides and API reference documentation? Yes. GraphStaff brings narrative developer guides and generated API references into one experience so readers can move from concepts to endpoints without changing products. #### Does GraphStaff support OpenAPI? Yes. GraphStaff supports OpenAPI 3.x, including multi-file specifications with external references, and turns the specification into endpoint pages with examples and code samples. #### How do authors edit and publish documentation? GraphStaff is being designed to support the workflows teams actually use. That may include repository-based publishing as well as a collaborative editor. We will choose the right authoring workflow with each early customer rather than forcing one model. #### Can private developer documentation use Ask AI and MCP? Yes. Identity and group permissions apply to the website, search, Ask AI, and MCP retrieval. Private pages are excluded before an answer or AI agent response is assembled. See it with your knowledge #### Make your documentation part of the developer workflow. Show us the guides, API definitions, and questions your developers work with. We’ll show you the human and AI agent experience together. --- ## Customer Knowledge Hubs | GraphStaff Canonical URL: https://graphstaff.ai/knowledge-hubs/ Summary: Branded, searchable customer knowledge hubs with grounded Ask AI, audience controls, and clear support escalation. Customer Knowledge Hubs ### Let customers have a live conversation with your product knowledge. Your public help center answers in plain language, grounded in your own articles and cited, so customers get the answer instead of a page of search results. The direct answer #### What is a customer knowledge hub? A customer knowledge hub is a branded self-service destination for how-to articles, troubleshooting, release notes, account guidance, and grounded answers. Customers can browse, search, or ask in natural language, while audience rules protect customer-only, partner-only, or plan-specific content across every retrieval surface. Self-service that serves #### Help customers finish the task, not browse a content archive. A clear front door #### Search, browse, or ask. Customers take the path that feels natural and get one version of the truth. help.yourcompany.com YYourCompany Help Release notes Customer knowledge hub ##### How can we help? Search for an answer... ⌘ K ↗ Getting started 8 articles → ◇ Account & billing 12 articles → ⌁ Troubleshooting 16 articles → ✦ Popular answers 10 articles → Help that travels #### Your hub gets its own MCP server. Give it to your customers. GraphStaff spins it up automatically when you publish. Customers connect it to ChatGPT or Claude, and their assistant answers product questions from your maintained help content instead of guessing. Your customer’s assistant help.yourcompany.com/mcp How do I turn off email notifications? mcp · search("email notifications") → Manage notifications · Help article Open Settings, choose Notifications, and switch off the emails you don’t want. Changes save automatically. Manage notifications · updated June 2026 One hub, different audiences #### Public help and protected customer knowledge live together. Publish open guidance beside authenticated collections for customers, partners, or plans. The same rules protect navigation, search, and AI answers. help.yourcompany.com Visitor Customer Getting started Public ✓ ✓ Troubleshooting Public ✓ ✓ Enterprise SSO setup Customers only – ✓ Premium onboarding playbooks Pro plan – ✓ Same rules in Navigation Search Ask AI Questions, answered #### Frequently asked questions. The short version of what teams usually want to know before they see GraphStaff in action. #### How is a knowledge hub different from developer documentation? A knowledge hub is organized around customer tasks and support questions rather than APIs and implementation. It prioritizes how-to articles, troubleshooting, account guidance, release notes, and clear escalation when self-service is not enough. #### Does GraphStaff replace our help desk? No. GraphStaff helps customers resolve questions before they become tickets, but it is not a ticketing system. When documentation is not enough, the hub can direct customers into your existing support workflow with better context. #### Can some help content require authentication? Yes. You can combine public help with customer-only, partner-only, or plan-specific collections. Access rules also apply to search and Ask AI, so protected content does not leak through an answer. #### Can customers ask questions in natural language? Yes. Grounded Ask AI searches and reads the knowledge hub before answering and can cite the relevant articles so customers can verify the answer or continue reading. See it with your knowledge #### Turn your support knowledge into customer confidence. Bring the questions your customers ask most. We’ll show you how the hub turns those questions into clear, grounded paths to resolution. --- ## Company Brain | GraphStaff Canonical URL: https://graphstaff.ai/company-brain/ Summary: A permission-aware Company Brain that gives Claude, ChatGPT, coding agents, and your AI infrastructure consistent company context through MCP. Company Brain ### Give every approved AI the same understanding of your company. Connect maintained internal knowledge to Claude, ChatGPT, coding agents, and your own AI infrastructure through one permission-aware MCP gateway. ASKED IN EVERY AI TOOL “How much parental leave do we offer?” CClaude New parents get 16 weeks of paid leave, usable anytime in the first year. People policy · updated May 2026 GChatGPT New parents get 16 weeks of paid leave, usable anytime in the first year. People policy · updated May 2026 ✦ One governed source, answered the same way everywhere. The direct answer #### What is a Company Brain? A Company Brain is a governed context layer for the AI tools a company approves. It makes maintained internal knowledge available through one permission-aware MCP gateway, so Claude, ChatGPT, coding agents, and internal AI infrastructure can answer with consistent company context without copying the same material into separate prompts. A shared context layer #### Update knowledge once. Use it with every AI. Stop maintaining a separate prompt library for every assistant. 01 YOUR COMPANY Trusted internal knowledge ▣ Product and processes How your company works ◇ Policies and handbooks The rules people rely on ◎ Customers and decisions Shared organizational context ⌁ Runbooks and operations What to do next GRAPHSTAFF Company Brain Live M NATIVE MCP SERVEROne governed gateway `search · list · grep · fetch` ◇ Identity ⌘ Groups ✓ Permissions Every request is grounded in the same company knowledge and filtered for the person or AI agent asking. 03 USE IT ANYWHERE Your approved AI tools C Claude Shared company context G ChatGPT Consistent grounded answers Coding agents Product and engineering knowledge ↯ Your AI infrastructure Workflows and internal assistants AM Alex Morgan Product · customers group GraphStaff policy ✓ Allowed context VISIBLE TO ALEX AND THEIR AI AGENT Product roadmap · Customer guidance NOT RETRIEVED Finance runbook · Executive notes Control who knows what #### Permissions travel with the question. Identity and groups are evaluated before retrieval, so each human or AI agent gets only the knowledge appropriate to that request. Questions, answered #### Frequently asked questions. The short version of what teams usually want to know before they see GraphStaff in action. #### How is Company Brain different from a shared prompt library? A prompt library copies context into separate instructions and becomes difficult to keep current. Company Brain keeps maintained knowledge in one governed graph and retrieves the permitted context when each AI tool asks for it. #### Does this require one specific AI model? No. The MCP interface keeps the knowledge layer independent from the model. Teams can use the same governed context with Claude, ChatGPT, coding agents, internal assistants, or their own AI infrastructure. #### How does access control work for AI agents? Every request carries an identity and group context. GraphStaff applies those permissions before retrieval, so an AI agent receives only the knowledge the person or service making the request is allowed to use. #### What kinds of company knowledge belong here? Useful starting points include product decisions, operating procedures, policies, customer context, technical runbooks, and internal handbooks, especially knowledge people repeatedly paste into prompts or explain in every new conversation. See it with your knowledge #### Stop explaining your company from scratch. Choose one high-value knowledge set and the AI workflow that needs it. We’ll show you what consistent, permission-aware context changes. --- ## Graph-Based Knowledge Retrieval | GraphStaff Canonical URL: https://graphstaff.ai/our-secret-sauce/ Summary: How GraphStaff's proprietary graph-based knowledge retrieval engine follows relationships to assemble complete, permission-aware context. Our Secret Sauce ### Knowledge retrieval that understands what connects to what. GraphStaff’s proprietary graph-based engine follows the structure and relationships in your knowledge to assemble context that is more complete, explainable, and useful. The direct answer #### What is graph-based retrieval? Graph-based retrieval represents knowledge as connected pages, concepts, policies, and explicit relationships instead of isolated text chunks. When someone asks a question, GraphStaff follows the relevant connections to assemble definitions, prerequisites, permissions, and exceptions into traceable context before a language model produces an answer. Why retrieval matters #### The answer is rarely in one chunk. A policy depends on a definition. An endpoint depends on an authentication rule. Useful knowledge is connected, but conventional retrieval treats every passage as if it lives alone. The knowledge graph #### Every question activates a neighborhood. The engine follows meaningful connections, so the context includes the definitions, policies, and exceptions the answer depends on. Source page Concept Policy PAGE Authentication CONCEPT Reader identity POLICY Customer access PAGE Account plans CONCEPT Entitlements PAGE Troubleshooting POLICY Internal only CONCEPT Escalation “What can this customer access?” A different retrieval model #### Similarity finds a passage. Relationships build an answer. Similarity-only retrieval GraphStaff retrieval Context Passages that sound like the query Relevant sources plus connected prerequisites and constraints Permissions Often filtered around the search layer Applied while traversing candidate knowledge Traceability A ranked list of text chunks Sources and relationships that explain why context was selected Questions, answered #### Frequently asked questions. The short version of what teams usually want to know before they see GraphStaff in action. #### Can GraphStaff show why context was selected? Yes. GraphStaff keeps the source pages and relationships used during retrieval, so the resulting context can be traced back to the definitions, prerequisites, policies, and exceptions that supported it. #### Is this the same as a vector search system? No. Semantic similarity can be one useful signal, but similarity alone does not express how concepts depend on, belong to, or constrain one another. GraphStaff uses explicit structure and relationships to guide retrieval and context assembly. #### Why does this improve an AI answer? Many questions require context spread across several related sources. Following known relationships helps the engine retrieve supporting definitions, prerequisites, policies, and exceptions instead of returning only the passage that sounds most similar to the question. #### How are permissions handled inside the graph? Access checks happen before protected knowledge becomes answer context. The engine filters candidate nodes and relationships using the requester’s identity and groups, so restricted material is not passed onward to the model. See it with your knowledge #### Bring us a question your current search gets wrong. A useful proof of concept starts with a difficult question and the knowledge needed to answer it. We’ll show you how the graph changes retrieval. --- ## Pricing | GraphStaff Canonical URL: https://graphstaff.ai/pricing/ Summary: GraphStaff pricing. Docs and Knowledge Hubs for $500 per month with everything included, and Company Brain starting at $80k per year with dedicated onboarding. Pricing ### One plan per product. Everything included. No tiers to decode, no features held hostage on a higher plan, and no free tier that quietly expires. Two products, two honest prices. The direct answer #### How much does GraphStaff cost? GraphStaff Docs and Knowledge Hubs cost $500 per month in one plan with every feature included. Company Brain starts at $80,000 per year on a one-year contract and includes dedicated implementation. There is no free tier, and a proof of concept uses your real knowledge before you buy. Docs + Knowledge Hubs #### One bundle. One price. Developer docs and customer knowledge hubs ship together as one product, so every audience answers from the same governed knowledge. $500 /month > We don't have the silly 3 tier option that only exists to get you to choose the middle tier. We don't play pricing games. Our plan is $500/month and that gives you everything. There is no other tier. ##### Everything means everything - Developer docs and API reference - Customer knowledge hubs - Grounded Ask AI - Authentication and permissions - MCP server, llms.txt, and per-page Markdown - Search and analytics Company Brain #### A newer product, rolled out deliberately. $80k /year Starting price, one-year contract. Right now we onboard Company Brain customers one at a time. It is a complex product that has to be implemented right the first time before it can help run your whole company. Every engagement includes dedicated onboarding and implementation from a real human. Questions, answered #### Frequently asked questions. The short version of what teams usually want to know before they see GraphStaff in action. #### Do you have special pricing for startups? No. We do not offer a startup discount. We do offer one discount, and it is available to everybody: if you help promote GraphStaff, we will lower your price. Reach out to our team to find out what that means. #### Do you have special pricing for non-profits? No. We do not offer a non-profit discount, for the same reason we do not offer a startup discount: one price for everybody keeps pricing honest. The one discount we offer is open to everyone, including non-profits: help promote GraphStaff and we will lower your price. Reach out to our team to learn more. #### Is there a free tier or a free trial? There is no free tier. Instead of a limited sandbox, we run a proof of concept with your real knowledge, so you evaluate GraphStaff on the questions your team actually gets asked. Request a demo and we will set it up. #### What does the $500 plan include? Everything. Developer docs, API reference, customer knowledge hubs, grounded Ask AI, authentication and permissions, the MCP server, llms.txt, and search analytics. There is no higher tier holding features back. #### $500 seems like a lot... Would it feel better if we listed it as $499? Sorry, but we respect our customers too much to play those games. The truth is, if you look at other similar platforms out there, once you get past the games the price is comparable. Jump up to the tier that actually matters, add up all of their hidden costs, all of their add-ons to get the features you actually want, add in the usage based charges on top of your subscription; do that and it will usually be about $500/month, if not more. With GraphStaff, you get a strong innovative product and a team that will support you without ever trying to upsell you, because there is no next tier to upsell you to. See it with your knowledge #### The demo is free. The pricing you already know. Bring us your docs and your hardest questions. We will show you what one governed knowledge graph does with them. --- ## GraphStaff for Support Teams | GraphStaff Canonical URL: https://graphstaff.ai/solutions/support-teams/ Summary: A branded customer knowledge hub with grounded, cited Ask AI, audience controls, and clean escalation into your existing help desk. Solutions ### Deflect the ticket, not the customer. A branded knowledge hub where customers search, browse, or ask, and every answer comes from help content you govern. The direct answer #### How does GraphStaff help support teams resolve questions? GraphStaff turns maintained support articles into a branded knowledge hub with search and grounded Ask AI. Customers receive cited answers before opening a ticket, protected content stays scoped to the right audience, and unresolved questions move into the existing help desk with clearer context. What support teams do with GraphStaff #### Answer once. Resolve it thousands of times. 01 ##### Publish the help center How-to articles, troubleshooting, and release notes in one branded destination. 02 ##### Let customers ask Grounded Ask AI answers from your articles and cites them, or refuses when the content is not there. 03 ##### Escalate with context When self-service ends, hand off to your help desk with the problem already narrowed. Audience controls included #### Plan-specific help without a second help center. Public help and customer-only collections live together. The same rules govern browsing, search, and AI answers. More solutions [API Teams](https://graphstaff.ai/solutions/api-teams/) [Technical Writers](https://graphstaff.ai/solutions/technical-writers/) [AI Platform Teams](https://graphstaff.ai/solutions/ai-platform-teams/) Questions, answered #### Frequently asked questions. The short version of what teams usually want to know before they see GraphStaff in action. #### Does GraphStaff replace our help desk? No. It resolves questions before they become tickets and hands the rest to your existing support workflow with better context. #### What stops the AI from making answers up? Grounded retrieval. The assistant reads your articles before answering, cites them, and refuses when the content is not there. #### Can customers on different plans see different help? Yes. Reader groups scope collections, and the same scoping applies to search and Ask AI. #### Who writes the articles? Your team, in MDX files in git. If your writers need a web editor today, GraphStaff is not the right fit yet. See it with your knowledge #### Fewer tickets. Calmer queues. Bring your top ten support questions. We will show you what governed self-service does to them. --- ## GraphStaff for API Teams | GraphStaff Canonical URL: https://graphstaff.ai/solutions/api-teams/ Summary: OpenAPI in, living reference out: synthesized examples, three-language samples, a safe try-it playground, and AI agent surfaces that respect permissions. Solutions ### Documentation your API can be judged by. OpenAPI in, living reference out: examples, code samples, a safe playground, and AI agent surfaces that respect your permissions. The direct answer #### How does GraphStaff turn OpenAPI into developer documentation? GraphStaff reads an OpenAPI 3.x specification and publishes endpoint pages with synthesized examples, code samples, version-aware search, and a safe request playground. The same reference is available to people through the website and to AI agents through MCP, llms.txt, and Markdown. What API teams do with GraphStaff #### Ship the reference with the API. 01 ##### Generate from the spec OpenAPI 3.x becomes endpoint pages with synthesized examples and three-language samples. 02 ##### Let developers try it A host-allowlisted proxy playground runs real calls without exposing keys. 03 ##### Version what you ship Docs versions with version-scoped search and assistant answers. 04 ##### Serve the AI agents too MCP, llms.txt, and Markdown twins publish beside the human site. Scoped by reader #### Partner endpoints stay partner-only. Reader groups scope the reference like any other page, and the scoping follows into search, Ask AI, and MCP. More solutions [Support Teams](https://graphstaff.ai/solutions/support-teams/) [Technical Writers](https://graphstaff.ai/solutions/technical-writers/) [AI Platform Teams](https://graphstaff.ai/solutions/ai-platform-teams/) Questions, answered #### Frequently asked questions. The short version of what teams usually want to know before they see GraphStaff in action. #### Which formats are supported? OpenAPI 3.x, including multi-file specs with external references. AsyncAPI and Swagger 2.0 are not supported today. #### Do you detect breaking changes? No. GraphStaff versions and publishes docs. It does not diff specs or generate changelogs today. #### Can each developer see their own keys and request logs? No. GraphStaff governs content and answers. It does not ingest API telemetry or manage developer keys. #### How do docs deploy? Push to git. CI validates, pull requests get preview URLs, and merge ships to production. See it with your knowledge #### A reference that keeps its promises. Send a spec, even a messy one. We will return a live reference you can judge. --- ## GraphStaff for Technical Writers | GraphStaff Canonical URL: https://graphstaff.ai/solutions/technical-writers/ Summary: One MDX source in git becomes the website, the search index, the AI answers, and the AI agent surfaces, with permissions intact on each. Solutions ### Write it once. It answers everywhere. One MDX source in git becomes the website, the search index, the AI answers, and the AI agent surfaces, with your permissions intact on each. guides/sso-setup.mdx merged · main `--- title: Single sign-on setup --- ## When SSO is required Enterprise workspaces can require SSO for every member. ` ◫ Website ⌕ Search ✦ Ask AI ⌁ AI agents The direct answer #### How does one MDX source serve people and AI agents? GraphStaff publishes one MDX source as a human-readable website, a search index, grounded Ask AI responses, MCP retrieval, llms.txt indexes, and per-page Markdown. Authors review changes through git, and the same version and audience rules follow the content across every published surface. What writers do with GraphStaff #### Your words, every surface, no copies. 01 ##### Author in MDX Twenty-two zero-import components: callouts, tabs, steps, code groups, and more. 02 ##### Publish through git Pull requests preview every change. Merge ships it. 03 ##### Answer through AI Ask AI reads your pages and cites them, so the words you wrote are the answer. 04 ##### Version without forking Versions partition navigation, search, and answers without duplicating files. One repository #### Internal notes and public guides, side by side. Group-scope a page and it disappears from every surface a reader cannot access: navigation, search, answers, and exports. More solutions [Support Teams](https://graphstaff.ai/solutions/support-teams/) [API Teams](https://graphstaff.ai/solutions/api-teams/) [AI Platform Teams](https://graphstaff.ai/solutions/ai-platform-teams/) Questions, answered #### Frequently asked questions. The short version of what teams usually want to know before they see GraphStaff in action. #### Is there a web editor? Not today. Authoring is MDX in git, and non-technical contributors work through pull requests. #### Can I control how AI uses my writing? The assistant only answers from published pages and always cites them. It refuses rather than improvises. #### Does versioning mean maintaining copies? No. Versions partition the same site, and search and answers stay scoped to the version the reader is on. #### What does GraphStaff cost? $500 per month for Docs and Knowledge Hubs. One plan with authentication, Ask AI, MCP, and analytics included. No tiers, no add-ons, no per-user fees. See it with your knowledge #### The single source, actually single. Bring one page you maintain in three places today. We will show the same page answering everywhere from one file. --- ## GraphStaff for AI Platform Teams | GraphStaff Canonical URL: https://graphstaff.ai/solutions/ai-platform-teams/ Summary: One governed MCP gateway feeds Claude, ChatGPT, coding agents, and your own AI infrastructure the same permission-aware knowledge. Solutions ### Context your AI agents can be trusted with. One governed MCP gateway feeds Claude, ChatGPT, coding agents, and your own AI infrastructure the same permission-aware knowledge. C Claude MCP CLIENT ACTING FOR alex@yourco · customers `tool callsearch("data retention policy")` ◇ Identity ⌘ Groups ✓ Permissions checked before retrieval ✓ Data retention policy PAGE · CUSTOMERS MAY READ ✓ Security overview PAGE · PUBLIC ✕ Internal legal notes NOT PERMITTED · NEVER RETRIEVED The direct answer #### How does GraphStaff govern MCP retrieval? Each MCP request carries the identity and group context of the person or service asking. GraphStaff checks those permissions before retrieving knowledge, excludes restricted pages from the context, and returns permitted sources through one model-independent gateway that works with Claude, ChatGPT, coding agents, and custom AI infrastructure. What platform teams do with GraphStaff #### Stop pasting context into prompts. 01 ##### Publish one gateway Every AI agent connects to the same MCP server instead of a per-team scrape. 02 ##### Enforce permissions in retrieval Tool calls carry the reader's groups, so AI agents receive only permitted content. 03 ##### Stay model-independent Swap the model or the AI agent. The knowledge layer does not change. 04 ##### Update once Fix the page and every AI agent answers correctly on the next call. Policy-checked retrieval #### The access decision travels with the answer. The policy engine that gates the website gates retrieval, so an AI agent acting for Alex sees exactly what Alex sees. More solutions [Support Teams](https://graphstaff.ai/solutions/support-teams/) [API Teams](https://graphstaff.ai/solutions/api-teams/) [Technical Writers](https://graphstaff.ai/solutions/technical-writers/) Questions, answered #### Frequently asked questions. The short version of what teams usually want to know before they see GraphStaff in action. #### Which AI agents can connect? Anything that speaks Model Context Protocol: Claude, ChatGPT, Cursor, and custom agents. #### Is this Company Brain? Company Brain is the product built on this capability for internal company knowledge, rolled out one customer at a time with dedicated implementation. #### Can AI agents write through the gateway? No. The MCP server is read-only by design. #### What does it cost? Documentation and knowledge-hub surfaces are included in the $500 per month plan. Company Brain starts at $80,000 per year on a one-year contract. See it with your knowledge #### AI agents are only as good as their context. Tell us which AI agents your teams run. We will show them answering from one governed source. --- ## Mintlify vs GitBook vs GraphStaff | GraphStaff Canonical URL: https://graphstaff.ai/compare/mintlify-vs-gitbook/ Summary: Compare Mintlify, GitBook, and GraphStaff across authoring, permissions, AI-agent access, retrieval, and pricing. Compare ### Mintlify vs GitBook vs GraphStaff Mintlify wins on developer polish. GitBook wins on collaborative editing. GraphStaff asks a different question: who gets which answer? Mintlify GitBook GraphStaff The direct answer #### Which documentation platform fits: Mintlify, GitBook, or GraphStaff? Choose Mintlify when developer polish and MDX components matter most. Choose GitBook when collaborative web editing is the priority. Choose GraphStaff when the answer itself must stay consistent across readers and AI agents, permissions must apply on every surface, and graph-based retrieval matters more than the broadest authoring feature set. The honest read #### Both are good. They are good at different things. ##### Pick Mintlify If your docs are written by engineers and judged by developers. MDX components, API references, and AI agent surfaces are first class. ##### Pick GitBook If marketing and support edit the docs every day. Its block editor and change requests are the best collaborative authoring in the category. ##### Pick GraphStaff If what you care about is the answer itself. One governed knowledge graph, permissions on every plan, and the same trusted answer for every reader, AI agent, and AI. Side by side #### Five questions that decide it. | Question | Mintlify | GitBook | GraphStaff | | --- | --- | --- | --- | | What it is | A polished developer-docs platform. MDX in git with a web editor, loved by developer-tool companies. | A docs and wiki platform with the category's best collaborative block editor. | A knowledge platform where docs, knowledge hubs, and AI answers come from one governed graph. | | Who gets which answer | Password protection on Pro. OAuth, JWT, and SSO reader authentication are Enterprise features. | Authenticated access starts on the Ultimate tier. | Password, JWT, and OIDC with reader groups on every plan. The same rules govern pages, search, Ask AI, and MCP. | | How answers are found | Hybrid search plus an AI assistant on Pro and above, metered in credits. | Built-in search plus an Ask assistant grounded in published content. 500 included answers on Ultimate. | Graph retrieval. Connected pages, concepts, and policies are assembled into cited, traceable context before the model answers. | | What AI agents get | llms.txt, Markdown twins, and MCP. | llms.txt, Markdown URLs, and MCP. | The same surfaces, with reader permissions enforced inside every one. An AI agent sees only what its reader may see. | | Pricing shape | Pro at $450 per month billed annually ($540 monthly), with AI usage metered in credits. | Ultimate at $249 per site per month plus $12 per user billed annually ($299 plus $15 monthly). | $500 per month. One plan, everything included. No per-user fees, no credits, no add-ons. | Competitor details checked July 2026 against official pricing and documentation. Sources: [Mintlify pricing](https://www.mintlify.com/pricing) [Mintlify MCP](https://www.mintlify.com/docs/ai/model-context-protocol) [GitBook pricing](https://www.gitbook.com/pricing) [GitBook authenticated access](https://gitbook.com/docs/site-access/authenticated-access) Why GraphStaff #### One graph. One set of rules. One price. ##### It remembers who is asking Permissions live in the graph, so the website, search, Ask AI, and MCP all give the same governed answer. ##### It follows the graph Retrieval walks connected pages, concepts, and policies instead of guessing from fragments, and cites where the answer lives. ##### It has one price Authentication is not an enterprise upgrade and AI is not a credit meter. Everything on this page is included. In context #### Every surface answers from the same source of truth. A GraphStaff site with search, Ask AI, and the API reference living over one governed source. docs.yourcompany.com Live Questions, answered #### Frequently asked questions. The short version of what teams usually want to know before they see GraphStaff in action. #### Aren't Mintlify and GitBook more mature? Yes. Both are older companies with bigger teams and features GraphStaff does not have. GraphStaff is deliberately narrower: it exists so that every reader, AI agent, and AI gets the same governed answer. If that is your problem, narrower is a feature. #### When is Mintlify the better fit? When your team lives in git, wants a large MDX component library, and will not need per-reader permissions before enterprise scale. #### When is GitBook the better fit? When non-technical teams edit the docs every day. GraphStaff has no web editor today; authors work in MDX and git. #### What does GraphStaff cost? $500 per month for Docs and Knowledge Hubs. One plan with authentication, Ask AI, MCP, and analytics included. No tiers, no add-ons, no per-user fees. See it with your knowledge #### Three good tools. One governed answer. Bring the question your team answers most often. We will show you what a governed answer to it looks like. --- ## Confluence vs Notion vs GraphStaff | GraphStaff Canonical URL: https://graphstaff.ai/compare/confluence-vs-notion/ Summary: Compare Confluence, Notion, and GraphStaff for team knowledge, permissions, AI access, grounded answers, and pricing. Compare ### Confluence vs Notion vs GraphStaff Confluence organizes the enterprise wiki. Notion makes the workspace delightful. GraphStaff asks a different question: what answer do your customers and their AI actually get? Confluence Notion GraphStaff The direct answer #### When should a team choose GraphStaff instead of Confluence or Notion? Choose GraphStaff when maintained knowledge must become a governed answer for customers, developers, and AI agents. Confluence is stronger as an enterprise wiki, and Notion is stronger as a flexible collaborative workspace. GraphStaff is built for cited retrieval, audience-aware delivery, and consistent answers across human and AI surfaces. The honest read #### Both are great workspaces. They shine at different things. ##### Pick Confluence If your teams live in Jira and need a mature enterprise wiki. Deep page trees, space-level permissions, and Atlassian integrations are its home turf. ##### Pick Notion If you want one flexible workspace to write, plan, and organize. Blocks, databases, and delightful editing make internal knowledge easy to build. ##### Pick GraphStaff If what you care about is the answer itself. One governed knowledge graph, permissions on every plan, and the same trusted answer for every reader, AI agent, and AI. Side by side #### Five questions that decide it. | Question | Confluence | Notion | GraphStaff | | --- | --- | --- | --- | | What it is | An enterprise team wiki tied to Jira and the Atlassian suite, built for internal documentation. | A flexible all-in-one workspace where docs, wikis, and databases live together and anyone can edit. | A knowledge platform where docs, knowledge hubs, and AI answers come from one governed graph. | | Who gets which answer | Space and page permissions for logged-in members. A public knowledge base needs an add-on, and it is not built for per-reader customer access. | Granular member permissions inside the workspace. Published pages go fully public, with no per-reader authentication on public content. | Password, JWT, and OIDC with reader groups on every plan. The same rules govern pages, search, Ask AI, and MCP. | | How answers are found | Wiki search plus Atlassian Intelligence and Rovo, scoped to Atlassian content and licensed seats. | Workspace search plus Notion AI, answering from pages the signed-in member can already see. | Graph retrieval. Connected pages, concepts, and policies are assembled into cited, traceable context before the model answers. | | What AI agents get | Content reaches AI through Atlassian's own assistants. There is no open llms.txt or per-reader MCP surface for an outside AI agent. | A public API and a Notion MCP server, scoped to what the connected account can access. | The same surfaces, with reader permissions enforced inside every one. An AI agent sees only what its reader may see. | | Pricing shape | Per-user pricing across Free, Standard, Premium, and Enterprise tiers, billed monthly or annually. | Per-user pricing across Free, Plus, Business, and Enterprise, with Notion AI billed per member. | $500 per month. One plan, everything included. No per-user fees, no credits, no add-ons. | Competitor details checked July 2026 against official pricing and documentation. Sources: [Confluence pricing](https://www.atlassian.com/software/confluence/pricing) [Confluence permissions](https://support.atlassian.com/confluence-cloud/docs/assign-space-permissions/) [Notion pricing](https://www.notion.com/pricing) [Notion MCP](https://www.notion.com/help/notion-mcp) Why GraphStaff #### One graph. One set of rules. One price. ##### It remembers who is asking Permissions live in the graph, so the website, search, Ask AI, and MCP all give the same governed answer. ##### It follows the graph Retrieval walks connected pages, concepts, and policies instead of guessing from fragments, and cites where the answer lives. ##### It has one price Authentication is not an enterprise upgrade and AI is not a credit meter. Everything on this page is included. In context #### Every surface answers from the same source of truth. A GraphStaff knowledge hub with search, Ask AI, and protected collections living over one governed source. help.yourcompany.com Live Questions, answered #### Frequently asked questions. The short version of what teams usually want to know before they see GraphStaff in action. #### Is this even the same category? Not exactly, and that is the point. Confluence and Notion are where teams write and store knowledge. GraphStaff is where that knowledge becomes a governed answer for customers, developers, and AI. If you are choosing what your readers and their AI actually get back, this is the comparison. #### When is Confluence the better fit? When your company runs on Jira and needs a deep internal wiki with space-level permissions. GraphStaff does not replace an enterprise team wiki. #### When is Notion the better fit? When you want one flexible tool to write, plan, and organize internal work. GraphStaff has no block editor or databases; authors publish from their own source. #### What does GraphStaff cost? $500 per month for Docs and Knowledge Hubs. One plan with authentication, Ask AI, MCP, and analytics included. No tiers, no add-ons, no per-user fees. See it with your knowledge #### Two great workspaces. One governed answer. Bring the question your customers and team ask most. We will show you what a governed answer to it looks like. --- ## Graph-Based AI Research | GraphStaff Canonical URL: https://graphstaff.ai/ai-research/ Summary: GraphStaff research on connected, permission-aware memory, graph-based retrieval, and grounded context for AI agents. AI Research ### We believe the future of AI is graph-shaped. Language models reason brilliantly and remember unreliably. We think the missing piece is structure: knowledge shaped as a graph of what connects to what, so a model can be grounded in something durable and inspectable. The direct answer #### Why does AI need graph-shaped knowledge? Language models can reason over supplied context, but they do not reliably remember a company’s changing facts, relationships, or permissions. A knowledge graph gives AI agents durable structure they can traverse, so retrieval can include connected prerequisites and exceptions while keeping every answer grounded in inspectable sources. The thesis #### Models reason. Graphs remember. Bigger models keep getting smarter, and that alone will not fix what a model does not know about your world. The gains left on the table are in the context: the right, connected, permission-aware knowledge delivered the moment a question is asked. A graph of pages, concepts, and policies is a memory a model can be held to. Where it ships #### Our retrieval engine is the research, in production. Everything we learn about graph-shaped context goes straight into GraphStaff’s graph retrieval engine: how to model relationships, how to walk them, and how to assemble cited, traceable, permission-aware context before a model answers. The product is where an idea has to survive contact with real questions. Open questions #### We are following the graph past retrieval. Graph retrieval is the first place these ideas ship, not the last. A few of the questions we are working on now. 01 ##### Memory that AI agents can walk A persistent, connected memory an AI agent traverses, instead of a context window it forgets between turns. 02 ##### Retrieval that reasons over relationships Following dependencies, prerequisites, and exceptions the way a person would, not matching text and hoping. 03 ##### Grounding you can verify Measuring whether an answer truly came from the source it cites, so trust is earned instead of assumed. A long-term bet #### This is where we’replacing it. Graph-shaped knowledge is the throughline from our research to the answer your customers and their AI get today. If that is a bet you want to see up close, come talk to us. --- ## Blog | GraphStaff Canonical URL: https://graphstaff.ai/blog/ Summary: Articles and guides from GraphStaff on grounded AI answers, developer docs, knowledge hubs, and making company knowledge work for people and AI agents. From GraphStaff ### The GraphStaff blog. Ideas and field notes for making company knowledge useful to every person and AI agent who asks. The direct answer #### What is the GraphStaff blog? The GraphStaff blog publishes articles, guides, and product announcements about grounded AI answers, developer documentation, customer knowledge hubs, and permission-aware company knowledge. Posts explain how one governed knowledge graph serves people and AI agents, share practical guidance for documentation teams, and record how GraphStaff features work and why they are built that way. July 29, 2026 #### Introducing GraphStaff One governed knowledge graph that answers customers, developers, employees, and their AI agents, with grounded and cited responses on every surface. [Read the article](https://graphstaff.ai/blog/introducing-graphstaff/) July 27, 2026 #### Retrieval that follows the connections Why GraphStaff retrieves knowledge by walking a governed graph of pages, concepts, and policies instead of ranking isolated text chunks. [Read the article](https://graphstaff.ai/blog/retrieval-that-follows-the-connections/) July 24, 2026 #### Your knowledge has a second audience now AI agents read your documentation too. How GraphStaff serves MCP, llms.txt, and Markdown so agents answer from your maintained content instead of guessing. [Read the article](https://graphstaff.ai/blog/your-knowledge-has-a-second-audience/) --- ## Introducing GraphStaff | GraphStaff Canonical URL: https://graphstaff.ai/blog/introducing-graphstaff/ Summary: One governed knowledge graph that answers customers, developers, employees, and their AI agents, with grounded and cited responses on every surface. [Home](https://graphstaff.ai/) [Blog](https://graphstaff.ai/blog/) Introducing GraphStaff Blog · By GraphStaff ### Introducing GraphStaff One governed knowledge graph that answers customers, developers, employees, and their AI agents, with grounded and cited responses on every surface. Published July 29, 2026 The direct answer #### What is GraphStaff? GraphStaff is an AI knowledge platform that turns your documentation, help articles, and internal knowledge into one governed knowledge graph. Customers, developers, employees, and AI agents ask questions and get grounded, cited, permission-aware answers through hosted sites, search, Ask AI, an automatic MCP server, and llms.txt. Every team we talk to has the same shape of problem. The knowledge exists. It lives in developer documentation, help articles, onboarding guides, and policy pages. What is missing is a reliable way for the people who need that knowledge, and increasingly the AI agents working alongside them, to get a trustworthy answer from it. GraphStaff is our answer to that problem. #### Three products, one graph GraphStaff publishes your knowledge as three products that share one foundation: [Developer Docs](https://graphstaff.ai/developer-docs/) for guides, API references, and the coding agents your users already run; [Customer Knowledge Hubs](https://graphstaff.ai/knowledge-hubs/) so customers find the answer before they open a ticket; and [Company Brain](https://graphstaff.ai/company-brain/) for the internal knowledge your employees and internal AI tools depend on. Each product stands on its own. You can start with one and expand when another use case becomes useful. #### Grounded means grounded Underneath all three sits the same [graph-based retrieval engine](https://graphstaff.ai/our-secret-sauce/). Instead of treating your content as a pile of disconnected text chunks, GraphStaff represents pages, concepts, and policies as a connected graph and follows those connections to assemble complete, traceable context before a language model writes a single word. Answers cite the pages they came from, so readers can verify instead of trusting blindly. Permissions travel with the knowledge. If a collection is customer-only, partner-only, or internal, those rules apply to navigation, search, and AI answers alike. #### Built for the second audience Your knowledge now has a second audience: AI agents. Every GraphStaff site ships with an automatic MCP server, llms.txt, and machine-readable Markdown for every page, so Claude, ChatGPT, and coding agents can answer from your maintained content instead of guessing. If that sounds like the knowledge platform your team has been assembling by hand, see [pricing](https://graphstaff.ai/pricing/) or request a demo. We would love to show you the magic. Publisher GraphStaff Document GS-BLOG-001 Version 1.0 Status Published Suggested citation: GraphStaff. “Introducing GraphStaff.” Version 1.0, published, 2026. [https://graphstaff.ai/blog/introducing-graphstaff/](https://graphstaff.ai/blog/introducing-graphstaff/) Copyright © 2026 GraphStaff. No open-content license is stated. See it with your knowledge #### Ready for answers your whole company can trust? Bring us your docs, help content, or internal knowledge. We will show you what one governed graph can do with it. --- ## Retrieval that follows the connections | GraphStaff Canonical URL: https://graphstaff.ai/blog/retrieval-that-follows-the-connections/ Summary: Why GraphStaff retrieves knowledge by walking a governed graph of pages, concepts, and policies instead of ranking isolated text chunks. [Home](https://graphstaff.ai/) [Blog](https://graphstaff.ai/blog/) Retrieval that follows the connections Blog · By GraphStaff ### Retrieval that follows the connections Why GraphStaff retrieves knowledge by walking a governed graph of pages, concepts, and policies instead of ranking isolated text chunks. Published July 27, 2026 The direct answer #### Why does GraphStaff use graph-based retrieval? Most retrieval systems rank isolated text chunks by similarity, which loses the structure that makes knowledge trustworthy. GraphStaff represents pages, concepts, policies, and their explicit relationships as a governed graph, then follows the relevant connections to assemble definitions, prerequisites, permissions, and exceptions into complete, traceable context before a language model answers. Ask a colleague a real question about your product and watch what they do. They do not skim five paragraphs that happen to share keywords with your question. They follow connections: the setup guide points at a prerequisite, the prerequisite mentions a plan restriction, the restriction has an exception for enterprise customers. The answer lives at the end of that walk. Most AI retrieval does not work that way. It slices documentation into chunks, scores each chunk against the question, and hands the top matches to a language model. The chunks are plausible. The connections between them are gone. #### What the graph preserves GraphStaff’s [retrieval engine](https://graphstaff.ai/our-secret-sauce/) keeps the structure. Pages, concepts, and policies are nodes; the relationships between them are explicit edges. When a question arrives, retrieval starts from the most relevant entry points and walks the graph, collecting definitions, prerequisites, and exceptions as it goes. The result is context that is more complete than any similarity ranking would assemble, and, just as important, explainable. You can see which pages contributed to an answer and why they were reached. #### Permissions are part of the walk Because retrieval traverses a governed graph, access rules apply during the walk, not as an afterthought. Content a reader cannot see is never assembled into their context, so a protected page cannot leak through an [Ask AI answer](https://graphstaff.ai/developer-docs/) or an agent’s MCP query. #### Why this matters for trust Grounded answers are only useful if your team can stand behind them. Citations tell readers where an answer came from. The graph tells you how it was assembled. Together they turn “the AI said so” into an answer you can audit, correct at the source, and improve with every edit. Publisher GraphStaff Document GS-BLOG-002 Version 1.0 Status Published Suggested citation: GraphStaff. “Retrieval that follows the connections.” Version 1.0, published, 2026. [https://graphstaff.ai/blog/retrieval-that-follows-the-connections/](https://graphstaff.ai/blog/retrieval-that-follows-the-connections/) Copyright © 2026 GraphStaff. No open-content license is stated. See it with your knowledge #### Ready for answers your whole company can trust? Bring us your docs, help content, or internal knowledge. We will show you what one governed graph can do with it. --- ## Your knowledge has a second audience now | GraphStaff Canonical URL: https://graphstaff.ai/blog/your-knowledge-has-a-second-audience/ Summary: AI agents read your documentation too. How GraphStaff serves MCP, llms.txt, and Markdown so agents answer from your maintained content instead of guessing. [Home](https://graphstaff.ai/) [Blog](https://graphstaff.ai/blog/) Your knowledge has a second audience now Blog · By GraphStaff ### Your knowledge has a second audience now AI agents read your documentation too. How GraphStaff serves MCP, llms.txt, and Markdown so agents answer from your maintained content instead of guessing. Published July 24, 2026 The direct answer #### How does GraphStaff serve AI agents? Every GraphStaff site automatically publishes machine-readable surfaces alongside the human ones: an MCP server that agents query directly, llms.txt files that summarize the site for language models, and clean Markdown for every page. Access rules apply to these surfaces too, so agents get grounded, permission-aware answers from your maintained content. For twenty years, documentation had one audience: people. You wrote for developers skimming at 2 a.m., for customers mid-task, for the new hire on day three. That audience is still here. But a second one has arrived, and it reads differently. Coding agents look up your API before writing an integration. Claude and ChatGPT answer questions about your product whether or not you participate in the answer. Support copilots draft replies from whatever version of your help content they can reach. The question is no longer whether AI agents will read your knowledge. It is whether they will read the version you maintain. #### Machine surfaces, published automatically Every [GraphStaff site](https://graphstaff.ai/developer-docs/) ships its machine audience surfaces automatically. An MCP server lets agents search and read your content directly; your customers can hand it to the assistant they already use. llms.txt gives language models a canonical summary of what your site covers. And every page is available as clean Markdown, without navigation chrome or layout noise. None of this is a separate publishing pipeline to maintain. Publish once, and the human site and the machine surfaces stay in step. #### The same rules for every reader An agent is a reader with an access level. GraphStaff applies the same permission rules to MCP queries and machine surfaces that it applies to navigation and search, so customer-only or internal content stays protected no matter who, or what, is asking. #### Answers you can stand behind When an agent answers from your maintained content, it cites real pages you control. When something is wrong, you fix the source once and every surface updates. That is a much better position than hoping a model’s training data got you right. Curious what your docs look like to an agent? Our own [llms.txt](https://graphstaff.ai/llms.txt) is right there. No humans allowed, but we will not tell. Publisher GraphStaff Document GS-BLOG-003 Version 1.0 Status Published Suggested citation: GraphStaff. “Your knowledge has a second audience now.” Version 1.0, published, 2026. [https://graphstaff.ai/blog/your-knowledge-has-a-second-audience/](https://graphstaff.ai/blog/your-knowledge-has-a-second-audience/) Copyright © 2026 GraphStaff. No open-content license is stated. See it with your knowledge #### Ready for answers your whole company can trust? Bring us your docs, help content, or internal knowledge. We will show you what one governed graph can do with it. ## Reuse and citation No public reuse license is stated on the marketing website. Use the canonical page URL when citing or linking to GraphStaff material. Product capabilities, availability, competitor details, and prices should be checked against the canonical page before publication. ## Generation note This file and llms.txt are generated from the built HTML during every production build. They should never be edited in the dist directory by hand.