Scribe MCP for AI. Find any process guide or knowledge page instantly.
Gemini Works with every AI agent you already use
…and any MCP-compatible client
How this MCP server connects to your AI agent
Scribe connects your company's process documentation library directly to your AI agent. You can ask natural language questions and get answers pulled from step-by-step guides (Scribes) or general knowledge pages.
It lets you search across all documents, filter by team, check recent changes, and audit what content exists in your organization.
What AI agents can do with Scribe Automation
Check scribe status
Verifies the connection status between your AI agent and Scribe's API endpoint.
Get documentation stats
Retrieves an overview count of all guides, pages, and documents in the system.
Get team documents
Fetches all documentation content specifically belonging to one defined team.
Your agent queries every guide and page in the entire Scribe organization simultaneously.
Your agent limits the search results to documents belonging only to a specified team, like HR or Finance.
Your agent retrieves a list of all teams configured within your Scribe organization for scoping searches.
Your agent finds documents created or updated within a specific date range, helping you track recent content additions.
Your agent pulls statistics on the total number of guides and pages stored in your knowledge base.
Ask an AI about this
Waiting for input…
What AI agents can do with Scribe MCP Server: 10 Tools for Documentation & Auditing
These tools let your AI client search across all documentation types, scope searches by team, check document statistics, and manage knowledge base content.
Make your AI actually useful.
Add this MCP to Claude, Cursor, or Windsurf and your AI stops guessing. It gets real tools to look things up, take action, and handle the stuff you keep doing by hand.
Start using Scribe on VinkiusCheck Scribe Status
Verifies the connection status between your AI agent and Scribe's API endpoint.
Get Documentation Stats
Retrieves an overview count of all guides, pages, and documents in the system.
Get Team Documents
Fetches all documentation content specifically belonging to one defined team.
List Teams
Returns a list of every distinct department or team configured in your Scribe...
Search Documents
Runs a general query across all Scribe guides and Knowledge Pages combined.
Search By Date
Searches for documents that fall within a specific date range, using YYYY-MM-DD format.
Search By Team
Limits the search scope to only include results from one specified team's documentation.
Search Pages
Restricts the search query to only Knowledge Pages, excluding formal step-by-step...
Get Recent Documents
Lists documents that were created within the last 30 days for immediate review.
Search Scribes
Runs a general query across all Scribe's dedicated procedural guides.
Security and governance baked right in.
Pick your AI client below to get set up. Just create a Vinkius account, subscribe, and you're instantly up and running. We handle the entire backend infrastructure, delivering out-of-the-box support for HTTPS Streamable, SSE, and OAuth2—zero messy routing required.
Claude AI
Open Claude Settings
Go to claude.ai, click your profile icon, then navigate to Customize → Connectors.
Add Custom Connector
Click the "+" button and select Add custom connector. Paste your Vinkius endpoint URL:
https://edge.vinkius.com/[YOUR_TOKEN_HERE]/mcp
Replace [YOUR_TOKEN_HERE] with your token
from cloud.vinkius.com. For OAuth-protected servers, expand
Advanced settings to add credentials.
Start a conversation
Open a new chat. The Scribe integration is available immediately — no restart needed.
Choose How to Get Started
Build a custom MCP for your own tools, or connect a ready-made integration from our catalog.
Build Your Own
Turn any API into an MCP. Import a spec, define Agent Skills, or deploy with MCPFusion.
- Import from OpenAPI, Swagger, or YAML specs
- Create Agent Skills with progressive disclosure
- Deploy to edge with MCPFusion framework
- Built in DLP, auth, and compliance on every call
- Real time usage dashboard and cost metering
- Publish to catalog or keep private
Make Your AI Do More
Start with Scribe, then connect any of our 5,100+ other servers whenever your AI needs more. One click, no limits.
- Use this MCP plus 5,100+ others, all in one place
- Add new capabilities to your AI anytime you want
- Every connection is secured and compliant automatically
- Track usage and costs across all your servers
- Works with Claude, ChatGPT, Cursor, and more
- New servers added to the catalog every week
VINKIUS INFRASTRUCTURE
Cloud Hosted
Managed infra
V8 Isolated
Sandboxed per request
Zero-Trust Proxy
No stored credentials
DLP Enforced
Policy on every call
GDPR Compliant
EU data residency
Token Compression
~60% cost reduction
Built on the Model Context Protocol (MCP) for Claude, ChatGPT, Cursor, and more
The Model Context Protocol standardizes how applications expose capabilities to LLMs. Instead of operating in isolation, your AI gains direct access to external platforms, live data, and real-world actions through secure, standardized connections.
This connection provides 10 powerful capabilities that interface natively with Claude, ChatGPT, Cursor, and other compatible AI platforms. No middleware. No custom integration required.
Finding internal processes shouldn't feel like deep-diving through SharePoint folders., Solved with Vinkius AI Gateway
Today, finding a clear process is painful. You open the intranet and are faced with dozens of departments, each with their own folder structure: 'HR Policies > PTO > Leave Forms.' Then you find one document that references three other documents, and finally, you realize the procedure you need was actually described in a wiki page filed under 'Miscellaneous' from two years ago. You waste 20 minutes clicking through tabs just to confirm if the process changed last quarter.
With this server, your agent connects directly to all those silos. Instead of navigating folders or guessing which document is authoritative, you simply ask: 'What's our current policy for PTO?' The system runs multiple checks—it knows to look at both formal guides (`search_scribes`) and general knowledge pages (`search_pages`). It gives you one answer, complete with context, without the clicking.
Scribe MCP Server: Get procedural documentation instantly.
Before this server, creating a step-by-step guide required someone to manually record every click and decision—a tedious process that often resulted in guides falling out of sync with reality. The only way to update them was to assign an owner who would then have to physically update the documentation across multiple platforms.
Now, you can use your agent to automatically generate or find these structured procedures. You tell it what happened, and it finds or builds the precise SOP. It’s a massive step up from just having static documents; you get actionable process knowledge right in your chat.
What your AI can actually do with this
Scribe connects your AI agent directly to all your company's process documentation library. It gives your agent access to everything—from formal step-by-step guides (called Scribes) to general knowledge pages. You can ask natural language questions, and your agent pulls answers straight out of the source material.
Before you start querying anything, you'll want to verify the connection status; use check_scribe_status to make sure your AI client is talking correctly to Scribe's API endpoint.
If you need a big picture view, run get_documentation_stats. That tool pulls an overview count of every guide, page, and document in the entire system. To see what’s fresh on the books, use get_recent_documents to list all content that was created within the last 30 days.
When it comes to searching, you've got options depending on how deep you need to dig. If you want your agent to check every single piece of writing—guides and pages combined—you run search_documents. But if you know you only care about formal procedures, use search_scribes for a targeted query across all dedicated procedural guides.
Conversely, if the info is general knowledge, not a step-by-step process, restrict your search to Knowledge Pages using search_pages.
To filter down how wide that net casts, you can control the scope by team or by date. If you need to know what's only in HR’s docs, first use list_teams to get a list of every department configured in Scribe. Once you have that list, your agent narrows the search using search_by_team, limiting results just to one specified team.
You can also pull all content belonging exclusively to a single team's documentation with get_team_documents. If you need to track changes over time, run search_by_date to query for documents that fall within a specific date range, using the YYYY-MM-DD format.
If your search criteria is hyper-specific—say, only looking at Finance's guides from last quarter—you can combine those searches. For instance, you could use search_by_team and then restrict that result set using a date range query.
019dd155-fd9c-73b6-b0d0-9eed388be0d6 
Here's how it actually works
The bottom line is: you talk to your knowledge base like talking to a person; the server executes complex search logic behind the scenes.
Subscribe to this server and provide your Scribe API key (Enterprise Grid or Global plan required).
Your AI client sends a natural language query (e.g., 'What's the process for expense reports?') to the MCP Server.
The agent runs the appropriate tool (like search_documents or search_by_team) and returns specific, context-rich answers to your chat window.
Who is this actually for?
Knowledge Managers who spend all day chasing document versions are prime users. Operations Leads need to monitor team compliance and documentation gaps. Support Teams use this when they're tired of reading 10 different manuals just to answer one customer question. This server hands you a single pane of glass over your entire corporate memory.
Uses list_teams and get_documentation_stats to map out content gaps, ensuring every department has up-to-date SOPs.
Directly uses the agent for customer queries, relying on tools like search_documents to provide step-by-step guides instantly.
Runs audits using get_recent_documents and search_by_date to ensure critical processes haven't fallen out of compliance or review cycles.
What Changes When You Connect
Audit content gaps quickly. Instead of guessing what docs exist, run get_documentation_stats to get a precise count of your entire documentation footprint.
Pinpoint information by department. Use search_by_team instead of broad searches when you know the answer lives in the 'Marketing' team's guides.
Stay current on processes. Need to know what changed last week? Run get_recent_documents or use search_by_date to focus only on new content since the last quarter.
Stop searching everywhere at once. The agent can run multiple queries—for example, first listing teams with list_teams, then running a targeted search using search_scribes for one of those teams.
Maintain process integrity. By having clear tools like search_by_team and search_documents, your agents deliver answers that are always scoped correctly, reducing hallucination risk.
See it in action
The Support Agent Needs a Guide
A customer calls with an obscure billing question. Instead of having the agent run a general search_documents query and getting 50 links, you use the agent to scope the search first by running list_teams (to confirm the Finance team) and then using search_by_team. This narrows results instantly, giving the support agent the one correct step-by-step guide.
The Operations Lead Needs a Compliance Report
Operations needs to verify if all departments have reviewed their onboarding guides. They run list_teams first, then loop through each team name and call get_team_documents. Finally, they check the total count with get_documentation_stats—all without manual clicks.
Finding a Stale Policy
You suspect an old policy document exists but you don't know when it was published. You use search_by_date to look at documents created before 2022-01-01, and then run search_pages to filter out the formal guides. This helps identify knowledge pages that might be outdated.
Comparing Guides vs Pages
You are writing a new policy but need to see what existing process documentation looks like. You first run search_scribes to see the formal, step-by-step guides, and then immediately follow up with search_pages to gather background context that isn't tied to a specific workflow.
The honest tradeoffs
The Wildcard Search
Asking the agent 'What do we know about onboarding?' and letting it default to search_documents. This returns everything—formal guides, background pages, team lists—making it impossible to tell what's actionable.
If you need a process guide, explicitly ask the agent to run search_scribes using 'onboarding'. If you just need general context, use search_pages.
Ignoring Team Scope
Searching for 'PTO policy' and getting results from both HR guides and Engineering documentation because your query didn't specify a team.
Always preface the search with a clear scope. For example, asking the agent to first run list_teams and then running search_by_team(HR) ensures all results are contained.
Assuming Recent = Current
Just looking at documents that were created in the last 30 days (get_recent_documents) and assuming they are fully vetted, leading to incorrect procedural advice.
Use get_documentation_stats first. Then use search_by_date to check for a specific historical range (e.g., 'last year') to ensure you aren't missing policies that haven't been updated recently.
When It Fits, When It Doesn't
Use this server if your primary need is retrieval and auditing. You have a massive, structured knowledge base, but the information is siloed by topic or department. It’s perfect for companies with dedicated Knowledge Managers who deal in process documentation (SOPs). Don't use it if you are trying to generate content from scratch; that's not its job. If your goal is simply 'write me a guide about X,' just let the agent write it. But if the goal is, 'find the approved, current step-by-step guide for X in our internal documentation,' this server runs the specific search_scribes tool you need. Remember: if you only want background context and not an actionable process, use search_pages. If you need to know where that context lives organizationally, start with list_teams.
If your content is unstructured (e.g., raw Slack chats or meeting transcripts) and hasn't been formalized into guides or pages, this server won't help until someone manually writes the Scribe guide first. This tool assumes you have a structured source of truth.
Questions you might have
How do I check if Scribe is connected to my agent? +
You run the check_scribe_status tool. It instantly reports whether your API key and connection are active, letting you know right away if you need to re-enter credentials or troubleshoot.
What's the difference between using `search_documents` and `search_scribes`? +
search_documents is a general query that pulls from all content types (guides, pages, etc.). Use search_scribes when you specifically need a formal, step-by-step procedural guide.
How do I find documents written by the Sales team? +
Use the search_by_team tool and pass 'Sales' as the parameter. This limits your search results only to content generated or managed by that department, making the answer clean and focused.
What if I want to know how much documentation we have? +
Run get_documentation_stats. This tool gives you a clear count of guides, pages, and total documents. It's fast and gives you a measurable overview of your knowledge base size.
How do I use `list_teams` to map out all the teams in my Scribe organization? +
The tool immediately returns a list of every team connected to your Scribe account. This helps you scope your search and understand department document ownership before running specific queries.
What is the correct format for using `search_by_date` to find old documentation? +
Search_by_date requires dates in YYYY-MM-DD format. You specify a start date and an end date, narrowing your search scope down to that exact window.
When should I use `search_pages` rather than the general `search_documents`? +
Search_pages limits results strictly to Knowledge Pages. Use this when you know the information is highly structured reference data, not a step-by-step guide.
How do I quickly track documentation activity using `get_recent_documents`? +
This tool pulls all documents created within the last 30 days. It's useful for auditing content freshness and identifying which guides need an update or review.
Can I search only for step-by-step guides? +
Yes! Use search_scribes to search exclusively for Scribe guides, or search_pages for Knowledge Pages only.
Do I need an Enterprise plan to use this? +
Yes, the Scribe Search & Retrieval API is available on Enterprise Grid and Global plans. Contact Scribe sales for API access.
Can I find documents by a specific team? +
Yes! First use list_teams to see available teams and their IDs, then use search_by_team or get_team_documents with the team ID.
We've already built the connector for Scribe. Just plug in your AI agents and start using Vinkius.
No hosting. No infrastructure. No complex setup.
All 10 tools are live and waiting.
You're up and running in seconds.
Gemini Vinkius gives your AI agents access to the full catalog of app connectors, all fully managed, secure, and enterprise-ready. One subscription, every tool you need.
Built, hosted, and secured by Vinkius. You just connect and go.