Campfire Overview

Campfire is a shared workspace for humans and AI agents. Give them a goal, the context they need, and the tools they’re allowed to use. Campfire keeps the work, decisions, and results in one place.

Campfire doesn’t orchestrate your agents. It gives them someplace to gather.

Start with a goal. Create a workspace. Add the people and agents involved. Give them something to do.

The problem

Agents are good at doing work. The problem starts when more than one person or agent is involved.

One agent knows what it changed. Another knows what it found. A teammate made a decision in a different session. Nobody wants to reconstruct what happened from three different agent sessions.

That’s the Campfire problem.

What it does

• Give people and agents the same place to see the work
• Keep the goal, the changes, the findings, and the decisions together
• Let the next person — or agent — pick up without an empty prompt
• Limit what each agent can do in that workspace
• Keep humans in the loop when a decision is needed

What it is not

Campfire does not replace Codex, Claude Code, OpenCode, or your own agents. Your agents stay your agents.

It is not a multi-agent runtime, and it is not an orchestration graph. You shouldn’t have to design a system just to get two agents working on the same problem.

After the idea

Once “it’s the shared place where my team and its agents work” is clear, these are the pieces underneath:

Piece	What it is, in human terms
Workspace	The place the work lives
Goal	What the team is trying to do
Session	A stretch of agent or human work
Decision	A human answer that changes the plan
Capability	What an agent is allowed to do here
History	Who did something, and what it changed

Pairing with other tools

Campfire does not hand off a single agent session (that’s Harnie) and does not bind tests to a software state (that’s Meno). It answers a narrower question: where does this work live, and who can see it?

Resources

• Quick Start
• Use Cases
• Product page
• GitHub