Spec Driven Development
MCP Server that facilitates spec-driven development workflows, not just Vibe Coding.
Description from the maker (original English)
Weigh a dependency's bundle cost from supplied module sizes, with shipped overlap netted.
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": {
"sizeverdict": {
"command": "npx",
"args": [
"-y",
"@reneturcios/sizeverdict",
"mcp"
]
}
}
}Run this in your terminal
claude mcp add sizeverdict -- npx -y @reneturcios/sizeverdict mcpPaste into .cursor/mcp.json in your project, or ~/.cursor/mcp.json
{
"mcpServers": {
"sizeverdict": {
"command": "npx",
"args": [
"-y",
"@reneturcios/sizeverdict",
"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": {
"sizeverdict": {
"type": "stdio",
"command": "npx",
"args": [
"-y",
"@reneturcios/sizeverdict",
"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 6, 2026
The registry name was verified with the GitHub account CloudCorpRecords, which owns the repo
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 (0.1.0) reached the registry 4 days ago.
Apps that support MCP, such as Claude Desktop, Claude Code, Cursor and VS Code. You also need Node.js, Docker on your computer.
In the same category: Spec Driven Development, Remote Desktop Commander, Fomu 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.
Kenya government form guides: checklists, draft letters, eCitizen, Huduma, timelines. Sample data.
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.
Adding a dependency is easy to agree to and hard to price. The headline size in a package listing is not what you actually pay: part of the new code may already be in your bundle, shipped by dependencies you already depend on. sizeverdict works out the real price from numbers you already have. You supply the candidate's modules with a path and a byte size for each one, the same facts for any dependencies you already ship, and optionally a budget for net new bytes. sizeverdict hands back the candidate's total bytes, the bytes that are already shipped, the net new cost, an overlap list, and a verdict band. It never downloads, installs, or measures a package; it computes from the supplied facts only.
A candidate module counts as overlapping when its exact path appears in at least one installed dependency's module list. Shared bytes are summed at the candidate's sizes, and each overlapping module is counted once, even when two installed dependencies carry the same path — the installed dependencies' own sizes for that path are never used. The overlap list holds one entry per overlapping module, sorted by path, each with the candidate's size and the distinct names of the installed dependencies carrying that path, sorted ascending. Net new bytes are candidate bytes minus shared bytes.
The verdict is read off the net new cost. Zero net new bytes is free_overlap. Up to 10240 bytes is negligible, up to 102400 is small, up to 512000 is moderate, and anything above that is large. Each threshold is inclusive at the top of its band: 10240 is still negligible, 10241 is small, and so on. When you supply budget_bytes, the result also reports whether the net new cost is over it; landing exactly on the budget is not over.
Malformed input — a missing candidate, a missing, empty, or non-string name, a missing or empty module list, a module with a missing or empty path or a negative or fractional byte size, the same faults inside an installed dependency, or a bad budget — is a validation error naming the offending path, with no partial result. Two modules with the same path inside one package, two installed dependencies with the same name, and a candidate whose name matches an installed dependency (candidate already installed: X) are also errors. Unknown extra fields are ignored.
npx @reneturcios/sizeverdict assess --input args.json
npx @reneturcios/sizeverdict --version
args.json contains the full arguments object:
{
"candidate": { "name": "left-pad-plus", "modules": [ { "path": "index.js", "bytes": 4000 }, { "path": "lib/util.js", "bytes": 9000 }, { "path": "lib/extra.js", "bytes": 2000 } ] }, "installed": [ { "name": "utils-kit", "modules": [ { "path": "lib/util.js", "bytes": 8500 } ] } ], "budget_bytes": 10000 }
The assessment is printed as JSON:
{
"candidate": "left-pad-plus", "candidate_bytes": 15000, "shared_bytes": 9000, "net_new_bytes": 6000, "overlap": [ { "path": "lib/util.js", "bytes": 9000, "also_in": ["utils-kit"] } ], "verdict": "negligible", "budget": { "budget_bytes": 10000, "net_new_bytes": 6000, "over_budget": false } }
Note the shared module contributes the candidate's 9000 bytes, not the installed copy's 8500. Exit codes: 0 on success, 1 on a runtime error (including an unreadable or invalid input file, or invalid input), 2 on a usage error.
npx -y @reneturcios/sizeverdict mcp
MCP client config:
{
"mcpServers": { "sizeverdict": { "command": "npx", "args": ["-y", "@reneturcios/sizeverdict", "mcp"] } } }
assess_bundle_cost — Weigh a candidate dependency's real bundle cost from caller-supplied module sizes: total bytes, the modules already shipped via installed deps (shared bytes), the net new cost, and a verdict band.candidate (object, required) — the dependency being weighed:name (string, required, non-empty) — candidate package namemodules (array, required, at least 1 item) — the candidate's shipped modules; each item:path (string, required, non-empty) — module path inside the packagebytes (integer, required, at least 0) — module size in bytesinstalled (array, optional) — dependencies already shipped, same shape as the candidate; each item has a name and a modules list in the same form. Absent means none are shipped.budget_bytes (integer, optional, at least 0) — net-new-bytes budget to check againstReturns {candidate, candidate_bytes, shared_bytes, net_new_bytes, overlap, verdict, budget}, where candidate echoes the candidate's name, overlap holds one {path, bytes, also_in} per overlapping candidate module sorted by path, verdict is one of free_overlap, negligible, small, moderate, or large, and budget is {budget_bytes, net_new_bytes, over_budget} or null when no budget was supplied.
docker run -i ghcr.io/cloudcorprecords/sizeverdict
docker run --rm -v "$PWD:/w" ghcr.io/cloudcorprecords/sizeverdict assess --input /w/args.json
npm install
npm run build # tsc -> dist/ npm test # tsx --test test/*.test.ts (pure local, no network) npm run dev # run the CLI from source
MIT — Copyright (c) 2026 Rene Turcios