# GitScrum Knowledge MCP > GitScrum Knowledge manages your team's collective memory in one place. It lets you build structured knowledge bases, store meeting decisions as persistent notes, and search across every single discussion thread, wiki page, and task without leaving your AI agent. ## Overview - **Category:** collaboration - **Price:** Free - **Tags:** agent-memory, wiki-documentation, team-discussions, knowledge-base, global-search, revision-history, note-taking ## Description Your agent needs more than just the current chat context to be useful. This MCP connects it directly to GitScrum, turning it into a central repository for everything your team has ever decided or documented. You can create structured notes that act as permanent memory, tracking revisions and decisions over time. Need process documentation? Build and maintain nested wiki pages right in your agent's workflow. When the team needs context, you don't have to copy-paste from Slack into a document; instead, you simply ask your AI client to search across every note, discussion, and wiki page simultaneously. Vinkius hosts this connector so you can connect it once from any compatible AI client, giving your agent full access to the whole catalog of organizational knowledge. ## Tools ### create_channel Sets up a new, dedicated communication channel for the team. ### get_channel Retrieves detailed information about a specific discussion channel. ### list_channels Provides a list view of all active discussion channels in the workspace. ### list_discussions Retrieves a list of every project's discussions available for querying. ### create_note_folder Creates organizational containers (folders) for grouping related notes by project or topic. ### list_note_folders Lists the various folders used to categorize and organize agent memory. ### rename_note_folder Changes the name of an existing note folder for better clarity. ### channel_messages Retrieves all messages from a specific discussion channel. ### search_channel_messages Searches for messages within a single specified communication channel. ### send_message Sends an update or finding to a designated team channel. ### create_note Generates a new note to store decisions or context, supporting full markdown formatting. ### delete_note Permanently removes an existing note from the knowledge base. ### list_notes Gets a list of all notes, perfect for reviewing stored context or decisions. ### move_note_to_folder Organizes a specific note by moving it into a designated folder. ### note_revisions Shows the full history of changes made to a single note, tracking its evolution. ### toggle_note_share Changes whether a note is visible and accessible to the rest of the team. ### update_note Appends context or refines an existing note without overwriting its original content. ### reply_to_message Posts a reply directly into an ongoing discussion thread. ### global_search Searches across all types of resources—tasks, wikis, notes, and discussions—at once. ### thread_replies Retrieves all replies associated with a single message within a thread. ### create_wiki_page Builds a new wiki page that can support nested documentation structures. ### delete_wiki_page Removes a wiki page, ensuring no data is orphaned. ### get_wiki_page Gets the full content of a wiki page for review. ### list_wiki_pages Retrieves an index of all wiki pages within the project documentation. ### restore_wiki_revision Reverts a wiki page to a specific, saved version from its past history. ### wiki_revisions Retrieves the complete revision history for any given wiki page. ### search_wiki Performs targeted searches specifically across the documentation content of wiki pages. ### update_wiki_page Modifies the content of a wiki page with new information. ## Prompt Examples **Prompt:** ``` Save a note with today's architecture decision about using event sourcing. ``` **Response:** ``` Note created: 'ADR-007: Event Sourcing for Order Service' with your decision rationale in markdown. I've placed it in the 'Architecture Decisions' folder. The note is private — say 'share this note' when you're ready to publish it to the team. ``` **Prompt:** ``` Search everything in our workspace for 'payment gateway integration'. ``` **Response:** ``` Found 11 results across your workspace: 📋 3 tasks (WEB-28, WEB-31, WEB-40), 📚 2 wiki pages ('Payment Flow Architecture', 'Gateway Comparison'), 💬 4 discussion messages in #payments channel, 📝 2 notes ('Payment Provider Research', 'Stripe vs Adyen Analysis'). Want me to open any of these? ``` **Prompt:** ``` Post an update in the #engineering channel about today's deployment. ``` **Response:** ``` Message posted to #engineering: '🚀 Deployment complete — v2.4.1 is now live in production. Key changes: payment gateway integration, performance optimizations, and 3 bug fixes. Full changelog in the wiki.' The team will be notified. Want me to create a wiki page with the detailed changelog? ``` ## Capabilities ### Search all resources Query tasks, wikis, notes, and discussions in a single search request. ### Manage persistent decisions Create, update, and archive structured notes to store context, meeting outcomes, or architecture decisions. ### Build documentation hierarchies Generate nested wiki pages for formal process guides and technical write-ups. ### Monitor team conversations List channels, read messages in threads, or search specific discussion streams. ### Track knowledge history View revision histories for both notes and wiki pages to see how information changed over time. ## Use Cases ### Onboarding a new engineer to legacy systems A new hire needs to understand why the payment gateway was changed from Stripe to Adyen. Instead of asking five people, they use `global_search` and instantly find the original notes (`list_notes`) detailing the decision, the wiki page explaining the process flow, and the relevant channel messages where the final sign-off occurred. ### Documenting a complex feature rollout The team finished implementing OAuth 2.0. A technical writer uses `create_wiki_page` to build the master documentation, then uses `create_note` to log all the specific implementation details and decision rationale. They can then share this knowledge by calling `toggle_note_share` when it's ready for review. ### Debugging a historical bug A senior engineer finds a subtle bug in the billing module. They check `wiki_revisions` to see if the process documentation changed recently and use `search_channel_messages` within the `#billing-support` channel to find the original conversation where that code path was last discussed. ### Managing agent context across weeks An AI developer needs their agent to remember complex internal APIs over several sessions. They use `create_note` with structured markdown, then regularly call `update_note` with new findings. This keeps the agent's 'memory' fresh and accurate without requiring human input every time. ## Benefits - Never lose context again. Instead of just searching chat logs, you can use the `global_search` tool to pull decisions from notes and wiki pages simultaneously. - Maintain a single source of truth for your processes. Use the ability to create nested wiki pages (`create_wiki_page`) so documentation always feels like one cohesive knowledge base. - Track how information changes over time. The `note_revisions` tool lets you see exactly when and why an architecture decision was updated, giving full accountability. - Keep your agent memory organized. You can create dedicated folders (`create_note_folder`) and use `list_notes` to review structured context before starting a new task. - Stay coordinated without constant meetings. Use the discussion tools—like listing channels or sending messages—to communicate findings directly where the team needs them. ## How It Works The bottom line is that your agent stops just chatting and starts acting as the team's single source of truth. 1. First, you subscribe to the GitScrum Knowledge MCP in the catalog and enter your required API tokens. 2. Next, you point your AI agent at this MCP. Your agent now treats all notes, wikis, and channels as structured data sources. 3. Finally, you ask a question like, 'What did we decide about payment gateway integration last month?' The agent executes the appropriate tools and returns synthesized answers from across all connected knowledge types. ## Frequently Asked Questions **How do I use `global_search` with GitScrum Knowledge?** Simply ask your agent a natural language question. The tool executes the search across all connected resources—notes, wikis, and discussions—and provides grouped results showing exactly where the information came from. **What is the difference between `create_note` and `create_wiki_page`?** `create_note` is for ephemeral but important decisions or meeting summaries that need revision tracking. Use `create_wiki_page` when you are building formal, long-form documentation with nested sections. **Can I track changes to a note using `note_revisions`?** Yes, the `note_revisions` tool tracks every change made to a note. This lets you see exactly what context was added or removed and when it happened, which is critical for audit trails. **How do I find messages from a specific thread using GitScrum Knowledge?** You can use `thread_replies` on an existing message ID. This pulls all subsequent replies into one view, keeping the conversation context intact and easy to review. **When I use `list_note_folders`, how do I organize my agent memory by project or topic?** You create folders to categorize your notes, keeping related context separate. This structure ensures that decisions and meeting notes stay grouped logically, preventing important information from getting mixed up across different projects. **How do I use `toggle_note_share` to control who sees my work?** This tool lets you manage the visibility of your notes. You can keep sensitive decisions private while publishing key findings only when and where the team is ready for them. **If I overwrite critical documentation, how do I use `restore_wiki_revision`?** You restore a wiki page to any previous version using this command. It’s your safety net, allowing you to recover accurate project documentation immediately after an accidental edit or loss of context. **How do I find specific information when I use `search_channel_messages`?** You target the search directly to a single discussion channel. This filters out all the noise from unrelated conversations, getting you straight to the relevant messages and status updates within that thread. **Can my AI agent use notes as persistent memory across sessions?** Absolutely — this is a core design goal. Use `create_note` to store decisions, context, or meeting summaries as markdown notes. Organize them into folders with `create_note_folder` and `move_note_to_folder`. Use `toggle_note_share` to publish findings to the team. Every edit is versioned via `note_revisions` so you can track how knowledge evolves. **Can the agent search across everything in my workspace at once?** Yes! The `global_search` tool performs a unified search across tasks, wiki pages, discussions, user stories, sprints, and notes. Results are grouped by resource type, so you instantly see where every mention lives. It's the fastest way to find anything in your workspace. **Can the agent participate in team discussions and reply to threads?** Yes. Use `send_message` to post updates to any discussion channel, and `reply_to_message` for threaded conversations. The agent can also create channels with `create_channel`, search message history with `search_channel_messages`, and review thread replies with `thread_replies` — enabling fully automated status updates and knowledge sharing.