Spec Driven Development
MCP Server that facilitates spec-driven development workflows, not just Vibe Coding.
Description from the maker (original English)
Durable agent sessions: start, watch, message and pause KeepState agents; approvals ask a person.
These three are never merged into one score: opinions, measurements and popularity are different things.
We have not written a plain-language summary for this one yet. Below are the facts we read from the registry.
Pick the AI app you use, then copy the settings.
You need
Which app do you use?
Pick your app above. We remember it, so next time the button says “Install in …”.
Open Claude Desktop → Settings → Developer → Edit Config, and paste this into claude_desktop_config.json
{
"mcpServers": {
"ks": {
"command": "npx",
"args": [
"-y",
"@keepstateai/cli",
"mcp"
]
}
}
}Run this in your terminal
claude mcp add ks -- npx -y @keepstateai/cli mcpPaste into .cursor/mcp.json in your project, or ~/.cursor/mcp.json
{
"mcpServers": {
"ks": {
"command": "npx",
"args": [
"-y",
"@keepstateai/cli",
"mcp"
]
}
}
}Your browser asks to open the app, then the app shows the settings and asks you to confirm. Key values are left empty for you to fill in.
Paste into .vscode/mcp.json in your project
{
"servers": {
"ks": {
"type": "stdio",
"command": "npx",
"args": [
"-y",
"@keepstateai/cli",
"mcp"
]
}
}
}Your browser asks to open the app, then the app shows the settings and asks you to confirm. Key values are left empty for you to fill in.
This one runs on your computer, and ChatGPT can only connect to servers on the internet. Pick another app.
Your choice stays on this device in a cookie; no account needed.
These settings are generated from the maker’s registry entry. If the README says otherwise, follow the README.
Runs on your computer; the tool list appears in your AI app after you install it.
We check information anyone can see; we have not read the code. Checks that do not apply are left out of the count. · Checked Nov 8, 2026
The registry name was verified against the domain keepstate.ai (the registry checks domain ownership)
We have not read this repo from GitHub yet
We have not read this repo from GitHub yet
This one runs on your computer; there is no internet endpoint
Runs on your computer and uses local settings instead
Runs on your computer; we can only read tool lists from internet endpoints
Compared with every active server in this category in the registry. Stars count only when the publisher owns the repo.
It passed 1 of 1 applicable basic checks. We have not read the code. Before adding a key or token, see what access the server asks for and use a key with only the permissions you need.
The registry does not list a required key, though some servers ask you to sign in on first use.
It runs on your computer. Your AI app starts it when needed, and data does not pass through the maker’s servers (unless it calls an online service itself).
The GitHub repo last changed –, and the latest version (0.1.38) reached the registry today.
Apps that support MCP, such as Claude Desktop, Claude Code, Cursor and VS Code. You also need Node.js on your computer.
In the same category: Spec Driven Development, Remote Desktop Commander, Filesystem MCP
Servers in the same category from other makers, sorted by stars.
Reading a card:★ GitHub starsiThree pillars: reviews · checks · usageiUpdated = latest version published to the registryi
MCP Server that facilitates spec-driven development workflows, not just Vibe Coding.
Remote Desktop Commander: use files, terminal and processes on your own computers from any AI app.
Filesystem MCP server for AI assistants.
Copied from the maker’s repo as written. PopNx MCP has not reviewed or endorsed it. HTML, images and scripts are removed. Links lead off this site.
Durable agent sessions: run an AI agent in a microVM, checkpoint it, kill it, wake it — and it resumes mid-sentence, files, memory, and running processes intact.
This is the thin client. Every verb terminates at the KeepState control plane; the client is a key, the building is keepstate.ai.
curl -fsSL https://keepstate.ai/install.sh | sh
brew install keepstateai/tap/ks
npm install -g @keepstateai/cli
Every path downloads the same public-CI-built binary and verifies its SHA256 against the release's SHA256SUMS before installing. Releases carry GitHub build-provenance attestations.
ks login # browser device flow
ks run # a durable session on the fleet ks checkpoint --stop && ks wake # the plug-pull, survived
ks doctor checks connectivity, token, and version. ks update self-updates (checksum-verified; a mismatch is refused).
A Cruise job is a goal, a check and a ladder of models. The fleet runs one attempt per rung from an exact save point, verifies the candidate in a verifier the agent cannot reach, and accepts only when the check passes. A failed rung escalates to the next one; after the last rung the job waits in your review queue. Nothing is accepted on a model's word.
ks cruise init --goal "make the inventory tests pass" # drafts .keepstate/cruise.json
ks cruise approve # locks it: .keepstate/cruise.lock ks cruise run # uploads the workspace, starts the job ks cruise status # rung, attempts, save points, spend, verdict ks cruise artifact # the accepted tarball, sha256-verified
init reads the repository: the check is the test command it finds (pytest, npm test, go test) or the one you name with --tests. A goal with no check is not a job; init stops and says so. It makes no request and never calls a model. approve records the manifest's digest; run refuses a draft that is not approved, a test or workspace changed since approve, and a workspace over 50 MiB (.git and .keepstate excluded). Defaults: 1800 s per attempt and a $2 spend ceiling (--spend); the ladder comes from the model table (ks cruise models, --ladder family:model,...). ks cruise logs, cancel and resume complete the set. Every --help makes no request.
rm -f $(command -v ks) && rm -rf ~/.config/keepstate
MIT licensed.