Transcribing your sessions
SoonSpeak instead of type. Familiar transcribes the session in real time and saves it to a searchable Foundry journal you can edit later.
Guides
Practical guides for running your Foundry table with Familiar. Plain method, no hype.
The thread through all of them is the same. Good AI D&D is good prep. You bring a published adventure with its maps, NPCs, and journals, structure it into Foundry, and the AI reads and runs what is on the page. It does not invent your story. You stay the author, whether it runs one scene or the whole DM side. These guides show you how.
Not playing D&D 5e? See what Familiar does on your system.
Install Familiar and connect an AI. Everything else builds on these two guides.
Install Familiar, connect the AI you already pay for, and let it run the published adventure you import. The on-ramp to an AI co-pilot in Foundry.
Connect the Claude, ChatGPT, or Gemini you already use to your Foundry game, over MCP. No API key, no per-token bill, just the assistant you know.
The path for the assistant you already use, and the model and connection choices behind it.
The most direct connection of the three: Claude Desktop and Claude Code speak MCP natively, no agent app in between. Your Anthropic plan drives the DM's side of a published adventure with no per-token bill.
ChatGPT reaches your Foundry game through Codex, the agent a ChatGPT plan already includes, and runs the DM's side of a published adventure from there. Prefer a key? Bring an OpenAI API key instead.
The cheapest way in: Antigravity connects Gemini over MCP on any Google plan, including a free account, or you paste a Google API key. Either route runs the DM's side of a published adventure.
Subscription or API key? MCP subscription versus a pay-per-token key, and which is cheaper for how your table actually plays.
Which model, and how hard should it think? A capable model with reasoning headroom is the second lever behind good prep. Turn thinking up for fiddly rules, down to keep replies fast.
Run the model on your own PC instead of a provider, so your world never leaves the machine and the bill moves from a token meter to your graphics card. Setup is one settings tab and one switch. Without a discrete GPU, expect minutes per turn.
The working parts of a running game: prep, characters, memory, combat, and voices.
Good AI D&D is good prep. Structure a published adventure, then hand the running to the AI.
Lay a published adventure into Foundry as journals, sheets, and a one-page outline, so the AI has the pages to run from.
Give an NPC who they are, what they know, what they want, and how they speak. Anchored to that, the AI voices them and fills the small edges itself.
A long campaign overflows any context window. Keep a searchable record in Foundry so the AI reads from your notes, not from a fading window.
Run combat with an AI that cannot cheat. Foundry's own dice enforce range, damage, and conditions, while you stay the DM.
Let the AI speak. Set a narrator voice and character voices, bring your own provider key, and Familiar voices each line in your browser as the scene plays.
Speak instead of type. Familiar transcribes the session in real time and saves it to a searchable Foundry journal you can edit later.
Solo and duet play, the rig behind my own table, and the rules pack that keeps any AI honest.
You know these modules cold and never get to play one. Bring a published adventure, take a single character, and Familiar runs the DM's side around you: the monsters, the NPC voices, the 2024 5e rules.
One PC, one player, a microphone, and an AI playing two heroes and every monster. The exact rig behind my own game, and the order to build it in.
A copy-paste rules pack that makes any AI a more disciplined Game Master. Roll only when it matters, run the adventure you prepared, never decide for the players. Every rule in it was paid for at a real table.
Looking for the full list? See everything Familiar can do.