# Amplenote MCP for AI Agents MCP > Amplenote connects your personal knowledge base and task manager directly into any AI client. You can use natural language prompts with your agent to search all notes, update project tasks, create new documentation, or organize ideas instantly. ## Overview - **Category:** productivity - **Price:** Free - **Tags:** task-management, personal-knowledge-base, scheduling, search-engine, workflow-automation ## Description Think of this MCP as hooking up your entire Amplenote vault—your research notes, meeting minutes, and to-do lists—right into whatever AI client you're using. Instead of switching tabs and manually copying text, you talk to the knowledge base directly through natural conversation. If you've ever had an idea stored in one note, a task list in another, and needed them both for a project plan, this lets your agent pull it all together instantly. It treats your notes not just as files, but as actionable data points. You can ask the AI to find every instance of a specific client name across months of research, or tell it to draft a summary based on an existing note structure. When you're ready for more connections like this, remember Vinkius is the #1 MCP Catalog; connecting all your services there keeps everything organized. The result is that your personal knowledge base stops being static storage and starts becoming a true operational co-pilot. ## Tools ### list_notes Retrieves a full list of all notes within the Amplenote workspace. ### get_note Fetches the complete content and metadata for a specific note using its unique ID. ### create_note Generates an entirely new note, allowing you to set both a title and Markdown body content. ### update_note Edits the title or markdown body of an existing note using its unique ID. ### delete_note Permanently removes a selected note from your vault by its unique ID. ### search_notes Performs a full-text search across every note and task stored in the Amplenote system using keywords. ### list_tasks Provides an overview of all tasks, showing their status, due dates, and which parent notes they belong to. ### get_task Retrieves the specific details for a single task item by its ID. ### create_task Adds an actionable new task to your system, complete with priority and due date tracking. ### update_task Changes a task's status, content, or other properties to reflect progress. ### list_tags Shows all the tags used across your notes and tasks, helping you understand your knowledge taxonomy. ### get_note_actions Discovers what operations can actually be performed on a specific note. ## Prompt Examples **Prompt:** ``` What are the current top-priority action items for the Alpha project? ``` **Response:** ``` **Project Alpha Action Items** Here’s a snapshot of everything marked as urgent: * **Task:** Finalize budget review for Q3. * **Due Date:** 2024-10-15 * **Assigned To:** Sarah M. * **Status:** Pending * **Task:** Draft initial architecture diagram. * **Due Date:** 2024-10-20 * **Assigned To:** Dev Team * **Status:** In Progress You can mark the budget review as complete if you've signed off on it. ``` **Prompt:** ``` Summarize all my notes about AI ethics and suggest three next steps. ``` **Response:** ``` Based on your vault, here’s a summary of your work: **Key Themes:** Bias mitigation (Notes: 4), Data privacy laws (Notes: 6), Future regulatory impact. **Suggested Next Steps:** 1. Write a formal comparison chart between GDPR and CCPA for the next meeting. 2. Create a new task to schedule an expert interview on AI bias by month-end. 3. Draft a 'Disclaimer' note summarizing current legal risks. ``` **Prompt:** ``` I need to add a quick update about the marketing budget change. ``` **Response:** ``` Got it. I found your main project planning note (ID: 8a93-11bb). I've appended the following section: *** **[Update] Marketing Budget Change:** The Q4 spend is shifting $5k from digital ads to content creation. Updated by AI Agent on 2026-06-12. *** The note has been updated successfully. ``` ## Capabilities ### Retrieve notes by content The agent can list every note in your workspace or pull the complete text and metadata for any specific document. ### Search across all stored data You can ask the AI to perform a full-text search, pulling up notes or tasks that match specific keywords anywhere in your vault. ### Create and edit documentation The agent will draft entirely new notes for you, or update existing ones by appending information or fixing content errors. ### Manage all tasks and to-dos You can list every pending task across your entire system, retrieve details on a specific item, or mark it as complete using the agent. ### Organize knowledge with tags The AI can read and analyze the tag structure of your notes to help you understand how your data is organized. ## Use Cases ### Writing an academic literature review A researcher needs to synthesize findings from 20 different sources. They ask the agent to `search_notes` for all mentions of 'Quantum Computing' and then use those retrieved notes to draft a structured outline, saving hours of manual reading. ### Closing out project phases A consultant finishes a client engagement. They ask the agent to `list_tasks` for that client and then iterate through them, using `update_task` to mark everything as 'Complete' and compiling a final status report. ### Onboarding new team members A project lead needs to get up to speed quickly. They ask the agent to pull all notes related to 'Client X Launch Checklist' using `get_note`, giving them an instant, comprehensive briefing. ### Ideation and content planning A writer has a vague concept. They use the agent to `create_note` with a placeholder title and then ask it to populate that note with related ideas pulled from their tag history, starting the project immediately. ## Benefits - Search notes like a pro. Instead of remembering where you saved a detail, just ask the agent to find content using `search_notes` across your whole library. - Keep tasks updated without leaving chat. You can use `create_task` or `update_task` directly in conversation, treating your knowledge vault as an active project dashboard. - Structure your ideas instantly. If you need a new document, the agent creates it for you using `create_note`, complete with title and initial Markdown body content. - Master your organization. Use `list_tags` to understand how every piece of information is categorized, making future searches much more precise. - Deep sync between AI and PKM. The MCP ensures that changes the agent makes—like deleting a note (`delete_note`)—are instantly reflected in Amplenote. ## How It Works The bottom line is that you talk naturally to your agent, and it executes complex actions across your entire Amplenote system without you ever leaving the chat window. 1. Install this integration in your active workspace. 2. Provide your Amplenote Personal Access Token; this gives the agent permission to read and write to your notes. 3. Use your AI client to converse with your ideas. Simply ask, 'Find all my incomplete tasks related to Q4 budgets,' and the agent handles the rest. ## Frequently Asked Questions **How do I get started using the Amplenote MCP with my existing workflow?** You connect your Amplenote account through Vinkius to any compatible AI client. Once connected, you simply start prompting your agent to perform actions like searching or drafting. **Can I use Amplenote MCP to track tasks for a team?** Yes, the MCP gives your agent access to all task data. You can list pending items and update statuses across various notes, keeping everyone aligned on priorities. **Does this help me find specific details in old research documents?** It does. Using full-text search, the agent finds content based on keywords anywhere in your vault—whether it’s a note or an old task description. It retrieves the context for you. **What if I need to organize my notes by topic or project?** The MCP lets your agent list and analyze your tag structure, helping you understand how your data is categorized. You can then ask it to perform actions only within those specific tags. **Is Amplenote MCP better than just using a search engine?** It’s much deeper because it doesn't just return links; the agent reads the context and performs actionable steps, like drafting a summary or creating a new task based on what it finds.