About AgentBrain Handoffs
AgentBrain Handoffs is a local Python delivery layer for teams of AI coding agents, routing work to the intended existing session and holding it while an agent is busy. Its bounded retries and delivery-state tracking are designed to avoid duplicate sends and interruptions.
Key features
- 01Session-specific handoffs
- 02Waits while recipients are busy
- 03Bounded retry and delivery tracking
- 04Work contracts with deadlines
- 05CLI, live page, and Python package
Use cases
- 01Route a coding task to a registered agent
- 02Queue work while an agent finishes a turn
- 03Track a task result or blocker
- 04Inspect and release a stuck delivery
MCP support
Each agent runs an MCP server with a fixed identity; its tools write handoffs to the inbox. The separate delivery engine sends them to agent sessions.
Configuration
Local server config is the same for every client. Copy it into your client’s MCP config file.
docker run -i --rm ghcr.io/willykeenan/agentbrain-handoffs:0.1.0Local servers need a runtime on your machine (e.g. Node.js 18+). Config file locations differ by client; see your client’s docs.
ghcr.io/willykeenan/agentbrain-handoffs:0.1.0stdioTools
Local servers expose tools only once running on your machine, and the maintainer has not provided a tool list.
Details
From the project READMEAgentBrain Handoffs
Exact, durable handoffs between AI coding agents. Each handoff is one new turn in the recipient's existing session.

Real screenshot of the demo; the data in it is invented sample data.
pip install .
handoffs demo
Open the URL it prints (http://127.0.0.1:8765/). Four simulated agents hand work to each other. Ctrl+C stops and deletes the demo data.
Why this exists
A team of coding agents already has sessions, inboxes and tools. What it lacks is delivery: a message that reaches the right session, waits if that agent is busy, is sent once, and is retried within a bound when the provider drops it.
Without that, handoffs get lost, delivered twice, or pasted into a session that is already in the middle of a turn. AgentBrain Handoffs is the delivery layer. It is a local Python package (Python 3.9+, standard library only) with a command handoffs, a live page, and an MCP server.
60-second quickstart
From a clone of this repository:
python3 -m venv .venv
.venv/bin/pip install .
.venv/bin/handoffs demo
Or, once the package is on PyPI:
pipx run agentbrain-handoffs demo
You should see:
AgentBrain Handoffs demo
Open http://127.0.0.1:8765/
Four simulated agents (Planner, Engineer, Reviewer, Writer) hand work to each other.
Nothing real is contacted. The demo data is temporary and deleted when you stop (Ctrl+C).
The page updates every few seconds. Handoffs move through Accepted → Working → Finished. Work is returned and closed on its own. --speed 4 makes the simulated turns finish faster.
To keep the demo on another port: handoffs demo --port 9876.
A longer walkthrough, including your own agents, is in docs/QUICKSTART.md.
Concepts
A handoff is a message addressed to one registered agent. The delivery engine turns it into exactly one new turn for that agent, at the session it had when the handoff was enrolled.
A work contract is a handoff with a title. The recipient returns a result (or a blocker); the sender closes it as accepted, revision, or blocked. A due time is optional; if it passes, the sender gets one non-waking reminder.
Delivery states, in plain words:
| State | Meaning |
|---|---|
| WAITING | Queued. Next engine pass will try to send it. A provider outage stays here too: the detail explains the hold and no attempt is used. |
| BUSY | The recipient is in the middle of a turn. This waits. |
| HELD | Blocked on purpose: the connection is blocked or delivery is paused. No attempt used. |
| UNAVAILABLE | The agent or its session is gone. Nothing is redirected. |
| OWNER_REJECTED | The session refused the write. One retry is allowed. |
| SENDING | The write is in flight. |
| UNCERTAIN | The write may have landed; the reply was lost, or an accepted turn could not be observed for 10 minutes. Observed, never resent. |
| ACCEPTED | The recipient's session took the turn. |
| RUNNING | The turn is in progress. |
| RETURNED | The turn finished. |
| FAILED | The turn failed or was interrupted, or the session refused it twice (then it was never delivered). |
| ACKNOWLEDGED | The recipient read the message; no separate turn is needed. |
| CANCELLED | Proven absent after enough history scans, or released by a person with handoffs release. Not resent. |
| DUPLICATE | Identical to a handoff still in progress. |
Guarantees
- Exact recipient. A handoff is delivered only to the agent it was addressed to, at the endpoint it had when enrolled.
- One write per attempt. The attempt is reserved before sending. A lost reply goes to UNCERTAIN and is reconciled by observation. A refused write is retried once; a second refusal ends the delivery as FAILED, so it never holds the recipient.
- Never interrupt. A busy recipient keeps its turn. A second handoff to the same agent waits as BUSY.
- Bounded resend. An errored turn is resent up to twice; an interrupted turn once, after ten minutes, and only while the work is still open. Each resend has a new request id.
- Honest release. An UNCERTAIN delivery that three complete history scans over an hour prove was never received is CANCELLED. An accepted turn nobody can observe for 10 minutes becomes UNCERTAIN, so the same rule applies. When you know how a stuck delivery ended,
handoffs release IDcancels it; nothing is resent either way. - Outages hold. A provider status-page guard can hold delivery without spending an attempt.
- Deadlines. Work past its due time alerts the sender once, as a non-waking notification.
Adapters
| Provider | Status | What it talks to |
|---|---|---|
demo | Stable | Simulated agents. Used by handoffs demo. |
command | Stable | An argv (no shell) or an HTTP POST. |
codex | Experimental | codex app-server JSON-RPC over stdio. |
claude-code | Experimental | Claude Code headless (claude -p --resume). |
See docs/ADAPTERS.md for settings, examples and the experimental caveats.
MCP
Each agent session launches its own MCP server. The identity is fixed at start, so a tool call cannot act as another agent.
Claude Code
claude mcp add handoffs -- handoffs mcp --agent me
Replace me with the agent id you registered (handoffs agent add me --provider claude-code ...).
Codex (~/.codex/config.toml)
[mcp_servers.handoffs]
command = "handoffs"
args = ["mcp", "--agent", "me"]
Cursor (MCP servers in Cursor settings)
{
"mcpServers": {
"handoffs": {
"command": "handoffs",
"args": ["mcp", "--agent", "me"]
}
}
}
Point every process at the same database with --db or HANDOFFS_DB. The engine (handoffs serve or handoffs run) is what actually delivers; the MCP server only writes to the inbox. Details: docs/MCP.md.
Project memory: the ContextLib plugin
ContextLib keeps a project's decisions, facts and lessons as plain Markdown files. Install it and point the same server at a library, and each agent gets the context_* tools (brief, search, get, record, supersede, review, capture, export, import, status) next to its handoff tools:
pip install "git+https://github.com/willykeenan/agentbrain-contextlib"
claude mcp add handoffs -- handoffs mcp --agent me --context-library "/Volumes/SSD/ContextLib"
$CONTEXTLIB_ROOT works in place of the flag. The plugin runs as the server's fixed agent id, so an agent's records are authored by exactly the agent that sent its handoffs. Without a library, the server offers handoff tools only.
CLI
handoffs init
handoffs agent add planner --provider demo
handoffs agent add builder --provider demo --name Builder
handoffs send planner builder "Add a health check" --title "Health check" --due-minutes 30
handoffs serve # engine + live page on http://127.0.0.1:8765/
handoffs inbox builder
handoffs read ID --as builder
handoffs return ID --as builder --summary "Health check is green"
handoffs close ID --as planner accepted
handoffs status --json
handoffs tick # one delivery pass (Codex and Claude Code turns wait for run or serve)
handoffs release ID # stop tracking a stuck delivery; never resent
handoffs run # engine only
handoffs mcp --agent builder # stdio MCP server
--db PATH or HANDOFFS_DB selects the database (default ./.handoffs/handoffs.sqlite3). --json is accepted on the commands that print records. Usage errors exit 2.
| Command | Purpose |
|---|---|
init | Create the database. |
agent add|list|remove | Register agents. --set KEY=VALUE is JSON when it parses. |
allow / block | Directed connections. |
config KEY VALUE | enabled, enabledAfter, connections, outageGuard. |
send FROM TO MESSAGE | Inbox message; --title makes it work. --key is idempotent. |
inbox / read / accept / return / close | Work lifecycle. --as or HANDOFFS_AGENT. |
status / tick / run / serve / mcp / demo | Inspect and deliver. |
release ID | Cancel one stuck delivery (for example UNCERTAIN) so its recipient is free. Nothing is sent. |
Security
The live page binds to loopback and refuses non-loopback Host headers (DNS-rebinding). POST /api/send needs the token in <db>.token (mode 0600). --public-demo exists only on handoffs demo (simulated agents, temporary data); that page cannot send and shows plain state sentences instead of adapter errors. MCP identity is the --agent you started with. Full model: docs/SECURITY-MODEL.md.
Roadmap
- Keep the demo and command adapters stable.
- Harden the Codex and Claude Code adapters as those app protocols settle.
- More status-page presets for
outageGuard. - Planned: Agent Rooms will use this delivery contract.
Run with Docker
The image is published at ghcr.io/willykeenan/agentbrain-handoffs for Apple silicon and Intel.
docker run --rm -p 8765:8765 ghcr.io/willykeenan/agentbrain-handoffs demo --host 0.0.0.0 --public-demo # the live demo
claude mcp add handoffs -- docker run -i --rm -v ~/.handoffs:/data ghcr.io/willykeenan/agentbrain-handoffs mcp --agent me
The MCP server writes to the handoff database in ~/.handoffs; run the engine (handoffs serve) on the machine where your agents live so it can deliver.
License
Apache-2.0. Copyright KE Studios.
Source: github.com/willykeenan/agentbrain-handoffs. Live demo: huggingface.co/spaces/willykeenan/agentbrain-handoffs.
AgentBrain Handoffs is the open delivery layer for AgentBrain, the agent runtime coming to Agent Rooms (agentrooms.io).
Comments
New to MCP?
From what the protocol is to installing your first server, these answers cover most first questions.
An open protocol that defines how AI applications connect to external tools and data in a uniform way. Clients talk to the model; MCP servers provide tools, resources and prompt templates. They exchange standard messages, so one server can be reused by many clients.