Stash MCP Server - Make AI IDEs even smarter with your team’s knowledge
by•
Stash MCP Server lets AI agents like Cursor, Claude, and Copilot access your team’s real context (code, docs, issues) so they can resolve tickets without endless prompting.
Just say "solve my assigned issue with the ID of …" - that’s it.
Replies
I saw MCP thingy blow up. But no startup actually betting on it with an interesting product. You're one of it.
alphaAI Capital
Will definitely be trying this out.
Great work! It’s really valuable how you’ve adapted the MCP concept to real-world use cases. This is exactly the kind of solution teams need to truly benefit from AI tools. 🚀
Stash
@ugurkeskekci Thank you so much Uğur! 🚀
Congratulations on the launch! The tools and ecosystem built around MCP are becoming increasingly robust!
Love the focus on “build, not search.” What integrations are live today beyond Cursor/Claude/Copilot, and is there a VS Code quickstart with a sample repo to try an end‑to‑end “resolve ticket” flow?
Stash
@marketing_doraverse Thanks! You can integrate Stash MCP server to any AI coding tool that supports MCP protocol. There is no open sample quick start repo as long as we provide Stash MCP Server to our paid users.
Big congrats on the launch! The MCP Server approach is spot on—letting Cursor/Claude/Copilot pull real context from code, docs, and issues so I can just say “solve my assigned issue with ID …”. How do you handle auth/permissions across private repos and trackers (GitHub/Jira/Confluence)? Any on‑prem option?
Stash
@annale Thank you so much! Yes, It is working exactly the way you said. We are handling auth/permissions with custom MCP token logic. Every user can create custom MCP token and put it into their config so that they can reach to related API. We have also on-prem support.
Do you expose an API so other agents (e.g., a Slack bot) can query the same context and post updates back to tickets?
Stash
@thanh_th_o_le Actually we don't expose an API for now but It can be a great addition to our product, thanks for the comment!
Love the “build, not search” mantra. Are you tracking metrics like MTTR, search time saved, or % of tickets auto‑resolved?
Stash
@duy_vo9 Thank you so much! Yes, we are tracking different metrics with our "Insights" feature!
Congrats on the launch! How does it ensure AI agents like Claude stay synced with evolving team docs and code without manual updates?