Airlock runs on your own machine — the client starts it, so there is no endpoint to ping. Last commit 18 Jun 2026.
Local daemon giving AI agents named locks, shared state, events, and a task queue over MCP.
What the code does
We read the source, 5 d ago · rules 995398eba69f
What this server is able to do. For an MCP server this is often the job itself — a terminal server runs commands because that is what it is for. Listed so you know what you are plugging in, not as an accusation.
unitPath := filepath.Join(unitDir, systemdUnitName)
Is this your server and something here is wrong? Tell us — corrections are free and do not require a plan.
This code can reach further than it looks
We found places where it runs commands, builds paths or queries from values it is given. None of that is a flaw by itself — it becomes one when the code changes, and code changes quietly between releases. We re-read it on every one.
Alternatives to Airlock
same job, measured the same wayLocal-first shared memory and task coordination for AI coding agents. Go daemon plus headless CLI.
Shared task layer for AI coding agents. One MCP surface: task_search, task_get, task_mutate.
Shared memory for coding agents and their teams: docs, epics, tasks and decisions over MCP.
Repo-native task continuity for AI coding agents: a reviewed task-state ledger served over MCP.
Tasks, comments, shared memory and handoffs for teams of people and AI agents, over MCP.
Open, local-first engram memory for AI agents: read, correct, delete; shared over MCP.
Shared rooms, messages, file locks and task claims for AI coding agents from any vendor.
Stateful, structured, safe shell sessions for AI agents over local, SSH, and Docker.
Airlock — questions
Answers built from our own checks of this server.