Doing Real Work with the DevStride MCP Server

Once your AI client is connected, you talk to your work in plain language. This page lays out the full range of what you can ask — grouped by what you're trying to get done — followed by a worked example of the assistant planning a sprint end to end, and example prompts by role.

What you can ask, by activity

Here's the range of what the assistant can do, grouped by what you're trying to get done. (You can also just ask "what can you do?" — it has a built-in catalogue.)

📋 Catching up

• "What should I focus on today?" — your open items, what's due, and what's blocking you.
• "What's the team working on right now?" — the current sprint, who has what, and what's stuck.
• "What did we ship this week?" — recently completed work, time logged, and comment activity.
• "What's coming up next?" — upcoming cycles, items due soon, and the backlog.

🔍 Finding things

• "Where's that item about topic?" — search by keyword, title, or number across every board.
• "Show me everything tagged label." — filter by labels, priority, or status.
• "What's under this Epic (or workstream)?" — hierarchy, dependencies, and who owns what.
• "Who's overloaded, and who has capacity?" — workloads, team rosters, and member lookups.

📚 Knowledge base

• "Search the knowledge base for topic." — search or browse articles, with the ranked results your readers get.
• "Draft an article about topic and file it under category." — create and update article content, and organize the category tree.
• "Publish this article" — also unpublish, archive, unarchive, and set which companies can see it.
• "Who last edited this article, and can we go back to the previous version?" — read an article's history and restore an earlier version.

✏️ Capturing work

• "Create a Story under parent with this description, assigned to person." — also Tasks, Bugs, Features, Epics, and comments.
• "Update these items — change priority, owner, or due date — in one shot." — bulk edits across many items at once.
• "Move this into the current sprint." — also reparent, change board, or archive.
• "Mark this done" — or reopen it; also edit titles, descriptions, custom fields, and watchers.

🎯 Prioritizing & sequencing

• "What should we ship first?" — rank items by priority, to the top or bottom of a status, or above/below a sibling.
• "What's blocking what?" — add or remove blocks, blocked-by, and related-to links.
• "Reorder this status so the Critical items are on top." — status, board, and parent ranking.

💬 Talking to the team

• "Comment on this item and @mention teammate." — post, reply, react, edit, or delete comments.
• "Who needs to know about this?" — find people by name or email, and manage watchers.

⏱ Tracking time

• "Start a timer on what I'm working on now." — also stop the running one, or check what's active.
• "I forgot to track yesterday — log 2 hours on item." — add, edit, or delete manual entries.
• "Show me the time logged against board last week." — review and report on time entries.

📊 Measuring delivery

• "What's our velocity over the last three sprints?" — velocity and throughput trends.
• "How's cycle time trending on the board board?" — cycle-time and lead-time metrics.
• "Give me a progress snapshot for the current cycle." — completed vs. remaining, at a glance.
• "Summarize time tracked by person this month." — time-tracking rollups for reporting.

✅ Spotting problems (when you ask)

• "What's missing on my items?" — descriptions, estimates, effort, or logged time.
• "Audit this board for sloppy tickets." — a board-level quality review.
• "This item looks like an orphan — what should its parent be?" — parent-candidate suggestions.

🗺 Big-picture planning

• "Draft a roadmap for next quarter." — create, update, and rank items on roadmaps (a roadmap is the Gantt).
• "Group these items into an initiative." — create initiatives, attach items, and rank them.
• "Add a launch milestone on the roadmap for the 15th." — set and adjust roadmap milestones.
• "Spin up the next two-week sprint in the cadence." — create and adjust cycles and their cadences.
• "What does our portfolio look like?" — an initiatives-and-roadmaps overview.

🛠 Your workspace

• "What labels, priorities, and statuses does this org use?" — org metadata, work types, and custom fields.
• "How does DevStride work?" — general help.

Even when there's no purpose-built tool for what you're asking, the assistant can reach the underlying DevStride API directly — so less common administrative and configuration tasks are within reach too, always within the permissions you already have.

Updating Gantt chart dates automatically — and keeping everything in sync

The assistant can set up dependencies and move dates, so you can keep a schedule current in plain language:

• "Make item B depend on item A." — links the two items.
• "Push item A's due date to next Friday." — changes an item's start or due date.

If your organization has Link Mode turned on, moving an item's due date through the MCP automatically adjusts the items that depend on it — the same behavior you get on the Gantt chart. See Link Mode for how dependent dates update.

A worked example: from backlog to roadmap

This is the kind of multi-step work the MCP makes possible — and a good model for how to drive it: the assistant does the analysis and proposes, you approve, and only then does it change anything.

Imagine you point the assistant at a sprint's worth of unestimated work:

1. Estimate the backlog. "Read the descriptions of every item in the New status and propose a time estimate for each. Wait for my approval before changing anything." It reads each item — and any linked pull request for extra context — and returns a table of proposed estimates with its reasoning.
2. Sequence the work. It identifies the dependencies between items — what has to happen before what — and proposes an order: the chains (for example, a shared-infrastructure item several others depend on), the largest items to spread across days, and any risks (like a design task that blocks a build task).
3. Strengthen a thin item. Where a description is too sparse to act on, it proposes an expanded version — a clear objective, a fuller description, and acceptance criteria — plus a revised estimate, for you to approve or edit. You stay the author; it just drafts.
4. Apply the plan. Once you approve, it updates each item's estimate and sets start and due dates per the schedule — in one pass, not one item at a time.
5. Build the roadmap. Finally, it groups the items into initiatives by theme and creates the roadmap (the Gantt timeline) with those initiatives — confirming the roadmap name, the initiative dates, and whether to attach items before it creates anything.

The result is a fully structured, dated roadmap built from your real backlog — drafted by the assistant, reviewed and approved by you at every step.

Examples by role

The same connection serves very different jobs. A few starting points:

Project / delivery manager

• "Summarize the current risks and blockers across the board board for our standup."
• "Show overdue items grouped by assignee."
• "Draft a roadmap for next quarter and group the work into initiatives."

Team member

• "What should I focus on today, and what's blocking me?"
• "Start a timer on the item I'm working on."
• "Create a Task under parent and assign it to me."

Service delivery leader

• "What did we ship this week, and what's at risk for the next release?"
• "Audit our active boards for items missing estimates, owners, or dates."
• "Compare planned vs. completed work this sprint."

Executive (CTO / VP)

• "Give me a portfolio overview — the initiatives and roadmaps in flight."
• "Where are the cross-team delivery risks right now?"
• "Draft a status update for leadership based on recent activity."

Working with sub-tasks

Your assistant can search across sub-tasks, list the sub-tasks of a parent, and create, update, or delete them. Item searches can also include items whose sub-task is assigned to you — so you can ask for "everything I'm working on, even via sub-tasks." Results are always filtered to what you have permission to read.

Working with personas

Personas are available through the API and MCP alongside person assignments. For example:

• "Show work assigned to Engineer that is still waiting for a person."
• "Assign this item to Jane and preserve Engineer if she holds that persona."
• "Show the personas each member can fill, in their fallback order."
• "Summarize recorded time by persona, keeping older unclassified time separate."

The same permissions and person/persona compatibility rules apply as in the app. A persona never grants item access. Catalog updates use the last-read catalog revision to prevent one administrator overwriting another's changes; saved personas are retired rather than deleted. See Managing personas.

Working with knowledge-base articles

Your assistant can work the Knowledge Base directly: find and search articles, read and write their content, move them through their lifecycle (publish, unpublish, archive, unarchive), set which companies see them, and manage the category tree.

It can also read an article's version history — who changed it and when — and restore an earlier version. A restore changes only the title and body, adds a new entry to the history, and deletes nothing, exactly as it does in the app. See Article history and restore.

Related

• Best Practices & Tips — habits that get you the best results.
• Setting Up Your Connection — connect your AI client.