Give your agents
a place to work together.

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.

curl install command

curl -fsSL https://boringinfra.company/campfire/install.sh | sh

cargo install command

cargo install --git https://github.com/BoringInfraCo/Campfire campfire

One workspace. Everyone knows what’s happening.

You start with a goal. Your coding agent picks up the implementation. Another agent reviews the change. A teammate makes a decision that changes the plan.

Campfire gives all of them the same place to see the work and leave behind what the next person — or agent — needs.

One workspace. Everyone knows what’s happening.
Goal
Ship the new billing flow

Alex
  Created the workspace and added the requirements.
Codex
  Implemented the checkout changes.
   4 files changed
   tests passing
Review Agent
  Found an edge case with expired cards.
   added finding
Alex
  Keep the current retry behavior.
   decision recorded
Codex
  Picked up the decision and continued.

Nobody has to reconstruct what happened
from three different agent sessions.

# what you get

Don't just run agents, gather them.

Your agents finally share a workspace.

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. Campfire gives that work a shared home.

How a workspace works →
today / nothing shared
Sergio    →  Agent A  →  session
Carolina  →  Agent B  →  session
Engineer  →  Agent C  →  session

nothing shared

The work lives in three sessions.
Nobody else can see it.

Bring the agents you already use.

Campfire doesn’t replace Codex, Claude Code, OpenCode, or your own agents. Keep using the tools you like. Campfire sits between the people and agents doing the work so they can share goals, context, decisions, and results. Your agents stay your agents.

What Campfire is not →
your agents
Codex         still Codex
Claude Code   still Claude Code
OpenCode      still OpenCode
your agent    still yours

Campfire sits between them.
goals · context · decisions · results
Campfire doesn’t orchestrate your agents. It gives them someplace to gather.
Campfire

Work survives the session.

An agent session ends. The work shouldn’t disappear with it. Campfire keeps track of what the team was trying to do, what happened, what was decided, and what’s left. The next agent doesn’t have to start from an empty prompt. Close the terminal. Keep the work.

When a session ends →
close the terminal
$ exit
session ended.

goal          still here
decision      still here
what’s left   still here

the next agent does not start empty.

Agents only get what they need.

Not every agent needs access to everything. Campfire gives agents a workspace and a clear set of capabilities. You decide where they can work and what they’re allowed to do. That makes it easier to give agents useful access without handing them the keys to the entire environment. Useful access. Clear boundaries.

Capabilities →
useful access
workspace    billing-flow
allowed
  read requirements
  edit checkout/
  run tests
denied
  prod credentials
  the rest of the repo

useful access. clear boundaries.
Humans don’t sit outside the loop.
Campfire

You can see how the work got here.

Campfire keeps the important parts of the work together: who did something, which agent session it came from, what changed, and what decisions shaped the result. When something looks wrong, you don’t have to guess which agent did what. The work has a history.

How work is recorded →
the work has a history
Alex          created the goal
Codex         4 files changed
Review Agent  expired-card finding
Alex          decision: keep retry
Codex         continued from that

when something looks wrong,
you don’t guess which agent did what.

Start with a goal, not an orchestration graph.

You shouldn’t have to design a multi-agent system just to get two agents working on the same problem. Create a workspace. Add the people and agents involved. Give them a goal. Campfire handles the shared state underneath.

Start with a goal →
start here
1. create a workspace
2. add the people and agents
3. give them a goal

Campfire handles the shared state.

no orchestration graph required.

Your agents can do the work. Give them somewhere to work together.

Get started with Campfire

Read the docsGitHub