Commons runs on your own machine — the client starts it, so there is no endpoint to ping. Last commit 18 Oct 2026.
Shared memory, work claims and handoffs so parallel AI coding sessions stop colliding.
What the code does
We read the source, 16 h ago · rules 995398eba69f
A value the model can set ends up inside a file or shell call. That is not a flaw by itself — for a terminal server it is the job — but it is where things go wrong when it is not.
r = subprocess.run(shlex.split(args.agent_cmd), input=prompt, text=True, capture_output=True,
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.
done=f"installed ~/Library/LaunchAgents/dev.commons.{name}.plist; {what} runs {every} (log {log})")
with open(os.path.join(out, f"{tag}.jsonl"), "w") as f:
r = subprocess.run(["git", "-C", top, *a], capture_output=True, text=True, timeout=3)
mod = __import__(mod_name)
out = subprocess.run(hook, shell=True, input=f"# {TITLES[sec]} ({bucket})\n\n{content}", capture_output=True,
${idn && team.length ? `<div class="ag-intro"><label class="ag-k" for="ag-intro-to">Introduce to</label><div class="ag-intro-r"><select id="ag-intro-to">${team.map(t => `<option value="${E(t.name)}">${E(t.display)}</option>`).join("")}</select>
env = dict(os.environ, COMMONS_WAKE_EVENT=payload["event"], COMMONS_WAKE_PAYLOAD=json.dumps(payload),
Is this your server and something here is wrong? Tell us — corrections are free and do not require a plan.
A tool parameter here reaches a dangerous call
That is not a flaw by itself — but it is where things go wrong when it is not the job. We re-read this code on every release. Watch it and you hear from us the day another one appears.
Alternatives to Commons
same job, measured the same wayShared memory for parallel AI coding agents. Local, free, with rooms and file claims.
Tasks, comments, shared memory and handoffs for teams of people and AI agents, over MCP.
Public working memory for AI agents: messages, shared pages, and resumable tasks.
Shared memory for coding agents: projects, reusable code, sessions and commits across machines
Coordination server for parallel coding agents: claims, contracts, notices, and memory notes.
Universal AI coding assistant memory — session handoff, SLM compression, and semantic retrieval.
Cross-harness communication and shared work-state for AI coding agents.
Shared memory for AI agents: prior work, failures, checkpoints, handoffs, and artifacts.
Commons — questions
Answers built from our own checks of this server.