Documentation

Configure Ollama, OpenAI, or other embedding providers

Run Ogham on Neon, self-hosted PostgreSQL, or any Postgres with pgvector

JetBrains AI Assistant added MCP client support in 2026.1. That brings every IDE in the JetBrains family – IntelliJ IDEA, PyCharm, GoLand, WebStorm, Rider, RubyMine, RustRover, Android Studio – into the same memory loop as Claude Code, Cursor, and the rest. Five minutes of setup; your assistant remembers across IDEs and across projects.

Prerequisites #

• A JetBrains IDE on 2026.1 or newer with AI Assistant enabled.
• Ogham installed and reachable from your shell (ogham --version should work). If you haven’t run ogham init yet, start with the setup wizard.

Configure the MCP server #

JetBrains AI Assistant has a UI for this, no manual JSON file editing required.

Set up Ogham persistent memory inside Kiro

Google Antigravity ships with native MCP server support via a single config file. Wire Ogham in once and the agent stores decisions, surfaces prior context, and stops re-explaining yourself across every session.

Prerequisites #

• Google Antigravity installed and signed in.
• Ogham installed and reachable from your shell (ogham --version should work). If you haven’t run ogham init yet, start with the setup wizard.

Configure the MCP server #

The config file lives at ~/.gemini/antigravity/mcp_config.json. Antigravity creates this file automatically (it’s empty by default until you add servers).

Run Ogham’s Postgres on a Proxmox LXC container.

Common issues and fixes

Typed edges add a queryable path layer on top of Ogham’s memory graph.

One-shot Linear import via the ogham CLI or the MCP tool.

Map Ogham’s compile_wiki, topic_summaries, and TLDR resolutions onto the consolidation pattern.