Skip to main content

Documentation Index

Fetch the complete documentation index at: https://docs.hopsule.com/llms.txt

Use this file to discover all available pages before exploring further.

MCP (Model Context Protocol) is an open way for AI assistants to use external tools and context. Hopsule exposes an MCP server so compatible assistants (Cursor, Claude Code, Copilot, custom hosts) read your decisions, memories, and project context as first-class tools instead of pasted text.

​
What it does

  • Exposes structured tools by domain: navigation (orgs, projects, active project), decisions, memories, tasks, context (Capsule- and project-scoped retrieval), and status helpers.
  • Applies role-based tool visibility from organization membership.
  • Expects a clear session project: agents should set an active project (or pass a project id) before project-scoped calls.
Typical flows include listing decisions, creating memories, and fetching context, with round trips to live project state.

​
When to use MCP vs other surfaces

Use MCP when…Prefer IDE add-on when…Prefer CLI when…
The client is an LLM agent or chat UI with MCP supportYou want diagnostics in the editor and sidebar browsingYou want a TUI, scripts, or push/status
The IDE add-on can use MCP alongside editing. Both reflect the same accepted decisions your team stored in Hopsule.

​
Security and auth

The hosted server uses Clerk-compatible auth for MCP (including OAuth where the client supports it). Treat tokens like production credentials: least privilege, no secrets in prompts.

Platform overview

How MCP fits the full product map.

IDE add-on

Editor-native MCP and advisory diagnostics.

Memories

What agents retrieve and cite.

Decisions

Accepted decisions and immutability.