K Kontinuo

Clean handoffs for coding agents

AI agents should know where the last one stopped.

Kontinuo is the continuity layer for AI coding agents. It turns agent sessions and workspace state into verifiable checkpoints, then serves them over MCP so the next tool can keep going.

Codex Claude Code Cursor Antigravity ACP editors

10

MCP tools for reading, writing, capturing, importing, and resolving handoffs.

0

Models run by Kontinuo. The agent reasons; Kontinuo records and serves state.

ACP

Protocol-level capture when agents run through an Agent Client Protocol editor path.

The gap

Agents do the work. Session boundaries erase the map.

01

Every agent keeps session history in a different place and shape.

02

Resuming means re-explaining context or hunting through old transcripts.

03

The next agent can repeat finished work, miss parked work, or trust stale state.

How it works

A handoff note good enough for a skeptical agent.

Kontinuo does not run agents, call models, or merge code. It never sends session data off the machine. It is the small local layer that keeps the handoff honest.

Capture

Transcript intent plus workspace facts

Kontinuo reads supported agent sessions, Agent Client Protocol streams, and git workspace state, then writes one normalized checkpoint for the current repo lane.

Verify

A checkpoint you can distrust safely

Each handoff records evidence, git HEAD, changed files, dirty state, and a workspace fingerprint so a reader can spot stale state before acting.

Resume

Served over MCP, readable by humans

The next MCP-capable agent calls handoff_read and gets the goal, stopping point, deferred work, verification notes, and the exact next action.

CLI capability

Install once. Let agents leave a trail.

The CLI handles setup, capture, inspection, and verification. The MCP server keeps the agent in flow: read the last checkpoint, work from it, then write the next one back.

1 Install the single Go binary.
2 Run setup to register the MCP server with your agent host.
3 Let agents call handoff_read and handoff_write while they work.

$ kontinuo version

out kontinuo dev

$ kontinuo log --cwd .

out No Kontinuo checkpoints found.

$ kontinuo setup --host claude --cwd . --config .mcp.json --store ./.kontinuo

out wrote MCP config for claude: /repo/.mcp.json

out store: /repo/.kontinuo

out next: restart or reload your agent so it starts `kontinuo mcp`

ACP support

Capture the handoff at the protocol layer.

Agent Client Protocol support lets Kontinuo observe editor-to-agent traffic directly. That means cleaner intent, fewer brittle transcript guesses, and a handoff the next agent sees before it starts work.

kontinuo acp --cwd . --inject compact -- codex

1

Protocol-level capture

ACP support observes the Agent Client Protocol stream, so Kontinuo can capture clear intent without betting the handoff on private transcript layouts.

2

Context at the right moment

When you run an agent through Kontinuo, opt-in context injection can place a compact handoff into the first prompt before the agent starts guessing.

3

Less brittle than scraping alone

Transcript adapters still help, but ACP gives editors and agents a cleaner path: the handoff travels through the protocol that already connects them.

The standard

Built to degrade, not break.

If a transcript adapter misses, Kontinuo falls back to workspace-only capture. If ACP is available, Kontinuo can observe protocol signals instead of leaning on private transcript layouts. If work is deferred, it stays parked.

Local-first store outside the repo by default.
Open JSON checkpoint schema with Markdown rendering.
Secret redaction across provider token formats.
Parked work stays parked until it is reopened.

Start local

Leave better notes for the next agent.

Add the MCP server to your agent host and let every session end with a checkpoint another tool can trust.