IntegrationUpdated
Connect Claude to your backlog
The Stride connector gives Claude your team’s delivery record: stories with their acceptance criteria, sprints, designs, documents, test cases and defects. It works in Claude on the web, in the desktop and mobile apps, and in Cowork, over the Model Context Protocol (MCP). You sign in with your Stride account and choose which projects Claude can reach.
What you need
- A Stride account in a workspace on a paid plan. Every paid plan includes MCP access; see pricing.
- A Claude account. On the Free plan you can add one custom connector. On a Team or Enterprise plan, an Owner adds the connector for the organization first.
How to connect it
The server URL is https://www.stride.page/api/mcp/stride. Use it exactly as written, with www.
On a Free, Pro or Max plan
- In Claude, open Customize → Connectors. In the desktop app, select Customize in the sidebar, then Connectors.
- Choose Add custom connector, name it Stride, and paste the server URL. Leave the authentication settings at what Claude detects (sign-in with OAuth, using Claude’s published identity); Stride needs no client ID or secret. Click Add.
- Click Connect. Stride’s sign-in opens: in the same tab on claude.ai, or in your browser from the desktop app.
- Sign in to Stride, choose the projects Claude can reach (below), and click Allow. You go back to Claude, where Stride shows as Connected.
On a Team or Enterprise plan
- An Owner opens Organization settings → Connectors, selects Add, then Custom (choosing Web if asked), pastes the server URL and clicks Add. This is done once for the whole organization.
- Each member opens Customize → Connectors, finds Stride with the Custom label, and clicks Connect to sign in with their own Stride account. Everyone keeps their own Stride permissions.
If Stride AI by Newlight Solutions appears when you search under Customize → Connectors → Discover, you can connect it from there with Connect to Claude instead (on Team and Enterprise, an Owner chooses Connect for your team first). It is the same server and the same sign-in. A connector named just Stride is a different product, made by another company.
What Stride asks when you sign in
- Connect it to: the project Claude’s work is recorded against, and the first one it reaches.
- Projects it can reach: that project plus any others you tick, or all your projects, including ones you join later.
- Claude acts with your role in each of those projects and never more, and everything it writes is labeled as the agent’s in Stride.
Use it in a conversation
In a chat, select +, then Connectors, and make sure Stride is on. Then ask for something your backlog holds, for example:
- “What’s still open in the current sprint, and who has it?”
- “Summarize ORB-526 with its acceptance criteria and any open defects.”
- “Write a test case for each acceptance criterion of ORB-526.”
- “File a high-severity defect against ORB-526: totals round down at checkout.”
- “Draft release notes for the last sprint and save them to the Library.”
Claude can ask before it uses a tool: choose Allow once or Always allow. Stride marks every tool that only reads as read-only and every delete as destructive, and Claude uses those marks for its prompts. To set them yourself, open Customize → Connectors, select Stride, and give each tool Always allow, Needs approval or Blocked.
What Claude can do
Stride exposes 131 tools. Each call is checked against your permissions in the project before it runs, and every change is recorded in Stride’s audit log.
| Area | What Claude can do |
|---|---|
| Plan | find, read and update stories, with their acceptance criteria, linked designs, documents and tests in one call; epics, sprints, releases, labels and story relations; bulk updates |
| Verify | test cases (one per acceptance criterion if you like), test runs and results, defects and quality gates |
| Design | generate UI mockups (HTML and CSS), architecture diagrams and BPMN processes for a story; read and link existing ones |
| Library | search and read documents such as meeting notes, PRDs, specs and retros; create documents and append to them |
| Optimize | processes, process insights and KPIs; legacy systems being modernized |
| Workspace | project members, comments and reactions, attachments, recent activity |
Generating a design, diagram or process uses your workspace’s AI credits and can take a few minutes; Claude checks on it until it is saved.
Access, privacy and disconnecting
- Only what you allowed. Claude reaches the projects you chose, with your role in each: in a project you can read but not edit, Claude can only read too.
- Labeled and logged. Comments and changes Claude makes are labeled as the agent’s, and every action is in the audit log.
- Disconnect from either side. In Stride, open Settings → Integrations and disconnect Claude under Connected apps; it loses access at once. In Claude, open Customize → Connectors, select Stride and choose Disconnect.
- Where requests come from. Claude calls Stride from Anthropic’s servers, not from your computer. How Stride handles your data is in the privacy policy and on Security & Trust.
Using Claude Code too
A connector you add on claude.ai is also available in Claude Code when you sign in there with the same Claude account. For coding work, the Claude Code guide covers the Stride plugin, whose /stride:next picks up your next story and /stride:done hands it back for review, and the agent keys that CI runs use.
If something goes wrong
- Claude can’t reach the server. Check the URL is exactly
https://www.stride.page/api/mcp/stride, withwww. - Stride shows Reconnect. The sign-in ended, for example because it was disconnected in Stride. Select Reconnect and sign in again.
- A project is missing. Claude reaches only the projects you chose. Disconnect, connect again, and tick the project on Stride’s sign-in screen.
- Sign-in stops at a plan limit. Each app you sign in uses one of your plan’s API or agent keys. Disconnect an app you no longer use under Connected apps, or move to a plan with more keys.
Frequently asked questions
How do I add Stride to Claude?
In Claude, open Customize → Connectors, choose Add custom connector and paste https://www.stride.page/api/mcp/stride as the server URL, then click Connect and sign in to Stride. On a Team or Enterprise plan an Owner adds it once under Organization settings → Connectors, and each member then connects with their own Stride account.
Does it work in the Claude desktop and mobile apps and in Cowork?
Yes. A connector you add is saved to your Claude account, so it is available in chats on the web, desktop and mobile, and in Cowork. It is also available in Claude Code when you sign in there with the same Claude account.
What can Claude see in Stride?
Only the projects you chose when you allowed it, and in each only what you can see yourself. Claude acts with your role in each project, so it cannot change anything you could not change, and everything it writes is labeled as the agent’s.
Which Stride plan do I need?
Every paid plan includes MCP access. Starter includes one API or agent key, and each app you sign in uses one, so on Starter Claude takes that one key; Pro and Enterprise raise the limit. See the pricing page for the full breakdown.
Do I need an OAuth client ID or secret?
No. Claude identifies itself to Stride with the client details Anthropic publishes for it, which Stride supports, so there is nothing to register or paste. Leave the authentication settings at what Claude detects.
How do I disconnect Claude?
In Stride, open Settings → Integrations and disconnect Claude under Connected apps; it loses access at once. Or, in Claude, open Customize → Connectors, select Stride and choose Disconnect.
Other integrations
- GitHubSource control · GitHub → Stride
- GitLabSource control · GitLab → Stride
- SlackCommunication · Stride → Slack
- Jira CloudProject management · 30-day two-way
- Claude CodeAI coding agent · MCP read/write
- OpenAI CodexAI coding agent · MCP read/write
- CursorAI coding agent · MCP read/write
- Webhooks & REST APIBuild your own · REST + webhooks
- How AI writes acceptance criteria (and where it fails)The honest map of where AI is dramatically better than humans at writing acceptance criteria, and the five places it confidently writes garbage. Plus the prompts that work.10 min read
- The connected delivery graph: one source of truth from PRD to prodMost teams ship software with five tools that don't talk to each other. The friction isn't any individual tool. It's the missing graph between them. This is the case for one connected graph.9 min read