CHAP Coordinator runs on your own machine — the client starts it, so there is no endpoint to ping. 355 installs a week from npm. Last commit 30 Oct 2026.
Auditable records of human decisions over AI agent work. Approvals, edits, overrides, escalations.
We read the source, 59 min 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.
proc = subprocess.run([node, os.path.join(_VENDOR, "render.js")], input=json.dumps(specs),
const probe = spawnSync(python, ["-c", "import chap_analytics"], { stdio: "ignore" });
path = os.path.join(directory, f"{name}.csv")
{ "name": "copy bomb exceeds size cap", "doc": {"s": "xxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxx…
Is this your server and something here is wrong? Tell us — corrections are free and do not require a plan.
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.
This server runs on your own machine — install it with the package manager and the client starts it for you. Package name taken from the official registry entry.
claude mcp add chap -- npx -y @brightbeamai/chap-coordinator-mcp
{
"mcpServers": {
"chap": {
"command": "npx",
"args": [
"-y",
"@brightbeamai/chap-coordinator-mcp"
]
}
}
}
[mcp_servers.chap]
command = "npx"
args = ["-y", "@brightbeamai/chap-coordinator-mcp"]
{
"mcpServers": {
"chap": {
"command": "npx",
"args": [
"-y",
"@brightbeamai/chap-coordinator-mcp"
]
}
}
}
{
"servers": {
"chap": {
"type": "stdio",
"command": "npx",
"args": [
"-y",
"@brightbeamai/chap-coordinator-mcp"
]
}
}
}
Optional settings from the registry entry, left out of the block above:
CHAP_DB_PATH (Path to a SQLite file. Without it the coordinator runs in memory and workspaces are lost when the client restarts), CHAP_PROFILES (Comma-separated CHAP profiles enabled on new workspaces. Defaults to Core plus review, whisper, deliberation, handoff, control, routing, modes and audit-scitt).
Governed browser hands for AI agents: classified actions, human approvals, replayable audit trails.
Record AI decisions to a tamper-evident audit ledger from any MCP-compatible agent.
Enables coding agents to coordinate edits, reviews, and escalations without file locking conflicts
Human-in-the-loop for AI agents over MCP: durable approvals with a hosted review page & audit trail
Human validation desk for AI agents: post work as cards, get a person's decision, carry on.
Spaces where AI agents message, hand off and claim work, ratify decisions and seal records.
Verifiable decision records for AI agents: signed receipts for calls to its own governed tools.
Desktop MCP workspace for AI agents: file edits, commands, Git status, resumable history sessions.
Answers built from our own checks of this server.