Discussions

Discussion management through MCP. Create and manage threaded discussion channels for project communication, decisions, and knowledge sharing.

Open Source — GitScrum MCP Server is open source under the MIT license. Available on npm and GitHub. Model Context Protocol server for GitScrum — Claude, GitHub Copilot, Cursor, and any MCP-compatible client full operational access to your project management stack.

The discussion tool provides 6 actions for creating and managing threaded discussion channels within your projects. Discussions are the structured alternative to scattered chat messages — they keep architectural decisions, design reviews, feature proposals, and team conversations organized and searchable within project context.

Every discussion lives inside a project, supports threaded comments, and can be opened or closed as topics are resolved. Your AI assistant can create discussions, post comments, track open threads, and close resolved topics — all through natural language in your IDE or AI client.

Actions Overview

Action	Purpose	Required Parameters
`list`	List all discussions in a project	`companyslug`, `project``slug`
`get`	Get full details of a specific discussion	`uuid`, `companyslug`, `project``slug`
`create`	Create a new discussion thread	`companyslug`, `project``slug`, `title`
`update`	Modify a discussion's title, description, or status	`uuid`, `companyslug`, `project``slug`
`comment_list`	List all comments on a discussion	`discussionuuid`, `company``slug`, `project_slug`
`comment_create`	Add a comment to a discussion thread	`discussionuuid`, `company``slug`, `project_slug`, `comment`

Listing Discussions

The list action returns all discussions within a project. You can optionally filter by status to see only open or closed threads. Each discussion in the response includes its uuid, title, status, creation date, and author — providing the context you need for follow-up operations.

Required Parameters

Parameter	Type	Description
`company_slug`	string	Workspace identifier (from the `workspace` tool)
`project_slug`	string	Project identifier (from the `project` tool)

Optional Parameters

Parameter	Type	Description
`status`	string	Filter by status: `open` or `closed`

You:  "List all discussions in the Backend project"
AI:   Calls discussion action=list → returns all discussion threads

You:  "Show me open discussions in the Mobile App project"
AI:   Calls discussion action=list with status="open" → returns active threads

You:  "Are there any closed discussions about the API redesign?"
AI:   Calls discussion action=list with status="closed" → AI filters by topic

Getting Discussion Details

The get action returns the complete discussion object — title, description, status, author, creation date, and metadata. This is the starting point when you need full context about a specific thread.

Required Parameters

Parameter	Type	Description
`uuid`	string	Discussion UUID (from `list` response)
`company_slug`	string	Workspace identifier
`project_slug`	string	Project identifier

You:  "Show me the details of the API design discussion"
AI:   Finds discussion by title → calls discussion action=get → returns full thread data

You:  "What was decided in the architecture review discussion?"
AI:   Calls discussion action=get → returns description and context

Creating Discussions

The create action starts a new discussion thread within a project. Every discussion needs a title; the description is optional but recommended for providing context to participants.

Required Parameters

Parameter	Type	Description
`company_slug`	string	Workspace identifier
`project_slug`	string	Project identifier
`title`	string	Discussion title — clear and descriptive

Optional Parameters

Parameter	Type	Description
`description`	string	Discussion body in Markdown format. Use this to frame the topic, list options, or provide background context.

Example Prompts

You:  "Start a discussion about API authentication strategy in the Backend project"
AI:   Calls discussion action=create with title="API Authentication Strategy",
      description with context about the topic

You:  "Create a discussion 'Database Migration Plan' with details about
       the PostgreSQL to CockroachDB migration timeline"
AI:   Calls discussion action=create with title and Markdown description

You:  "Open a thread about the mobile app launch checklist"
AI:   Calls discussion action=create with title="Mobile App Launch Checklist"

Updating Discussions

The update action modifies an existing discussion's title, description, or status. Only the fields you specify are changed — everything else remains untouched. Closing a discussion signals that the topic is resolved.

Required Parameters

Parameter	Type	Description
`uuid`	string	Discussion UUID
`company_slug`	string	Workspace identifier
`project_slug`	string	Project identifier

Optional Parameters

Parameter	Type	Description
`title`	string	Updated discussion title
`description`	string	Updated discussion body (Markdown)
`status`	string	Change status: `open` or `closed`

You:  "Close the API authentication discussion — we decided on OAuth2"
AI:   Finds discussion → calls discussion action=update with status="closed"

You:  "Update the migration discussion title to 'Database Migration Plan v2'"
AI:   Calls discussion action=update with title="Database Migration Plan v2"

You:  "Reopen the deployment strategy discussion"
AI:   Calls discussion action=update with status="open"

Discussion Comments

Comments form the threaded conversation within a discussion. Each comment supports Markdown formatting, enabling rich content including code blocks, links, and formatted text.

Listing Comments

The comment_list action returns all comments on a discussion in chronological order.

Parameter	Type	Description
`discussion_uuid`	string	Discussion UUID (required)
`company_slug`	string	Workspace identifier (required)
`project_slug`	string	Project identifier (required)

You:  "Show all comments on the architecture discussion"
AI:   Calls discussion action=comment_list → returns threaded comments

You:  "What has the team said about the API redesign?"
AI:   Finds discussion → calls comment_list → summarizes the conversation

Creating Comments

The comment_create action adds a new comment to an existing discussion. Comments support full Markdown — code blocks, bullet lists, links, and formatted text.

Parameter	Type	Description
`discussion_uuid`	string	Discussion UUID (required)
`company_slug`	string	Workspace identifier (required)
`project_slug`	string	Project identifier (required)
`comment`	string	Comment body in Markdown format (required)

You:  "Add a comment to the API discussion: 'After reviewing benchmarks,
       I recommend we go with GraphQL for the public API'"
AI:   Calls discussion action=comment_create with Markdown comment

You:  "Post the pros and cons of each database option to the migration thread"
AI:   Calls discussion action=comment_create with formatted Markdown list

You:  "Comment on the architecture discussion with notes from today's meeting"
AI:   Calls discussion action=comment_create with meeting notes in Markdown

Threaded Communication Workflow

Discussions work best when combined with other MCP tools to create a structured decision-making process:

1. Start the conversation

Create a discussion with clear framing and background context.

You:  "Create a discussion 'Auth Strategy for v3 API' in the Backend project.
       Include context about our current JWT setup and the need to support OAuth2."
AI:   Creates discussion with title and detailed Markdown description

2. Gather input

Team members add comments with their analysis and recommendations.

You:  "Add a comment with my analysis: OAuth2 + PKCE for mobile, API keys
       for server-to-server, session cookies for the web dashboard"
AI:   Posts formatted comment with the technical recommendation

3. Review the thread

Summarize comments to synthesize team input.

You:  "Summarize the comments on the auth strategy discussion"
AI:   Calls comment_list → generates a concise summary of all positions

4. Resolve and act

Close the discussion and create tasks based on the decision.

You:  "Close the auth strategy discussion and create tasks for implementing
       the OAuth2 approach we agreed on"
AI:   Updates discussion status to closed → creates tasks from the decision

Context Auto-Resolution

The discussion tool supports automatic context resolution. If your AI assistant already knows the workspace and project from a previous call in the conversation, it carries that context forward. You can say "list discussions" without specifying the project every time, as long as the context was established earlier.

Next Steps

Comments: Manage comments on tasks for operational collaboration.
Wiki: Build project knowledge bases for long-term documentation.
Tasks: Create tasks from discussion outcomes.
Quick Start: Set up the MCP server if you haven't already.