Spec Driven Development
MCP Server that facilitates spec-driven development workflows, not just Vibe Coding.
Description from the maker (original English)
Executable architecture governance: an unevaluated rule is reported as a failure, never a pass.
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": {
"evolith": {
"command": "npx",
"args": [
"-y",
"@beyondnet/evolith-mcp"
]
}
}
}Run this in your terminal
claude mcp add evolith -- npx -y @beyondnet/evolith-mcpPaste into .cursor/mcp.json in your project, or ~/.cursor/mcp.json
{
"mcpServers": {
"evolith": {
"command": "npx",
"args": [
"-y",
"@beyondnet/evolith-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": {
"evolith": {
"type": "stdio",
"command": "npx",
"args": [
"-y",
"@beyondnet/evolith-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 7, 2026
The registry name was verified with the GitHub account beyondnetcode, 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 (1.3.3) reached the registry 3 months ago.
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, FlyDocs
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.
Spec-driven development workflow: your agent works issues, sessions and the status lifecycle.
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.
Bilingual navigation: Versión en Español
Decouples the MCP server from the CLI. It is a first-class product that exposes the MCP tools as a Gateway talking to @beyondnet/evolith-core (the reusable business-logic layer), instead of shelling out to CLI subprocesses.
sequenceDiagram
participant Agent as "AI Agent (Cursor, Claude Desktop, Custom)" participant Gateway as "MCP Gateway @beyondnet/evolith-mcp" participant Core as "Business Logic @beyondnet/evolith-core" participant FS as "File System" participant Git as "Git"
Note over Agent,Gateway: Transport: stdio (local) or Streamable HTTP (remote)
Agent->>+Gateway: tools/call { name: "evolith-validate", args: { path: "/repo" } }
Gateway->>Gateway: 1. Generate correlationId (evl-xxx) Gateway->>Gateway: 2. Lookup tool in ToolRegistry Gateway->>Gateway: 3. ABAC authorization check Gateway->>Gateway: 4. Start timing + structured log (Pino)
Gateway->>+Core: ValidateSatelliteUseCase.execute({ satellitePath }) Core->>+FS: Read evolith.yaml, rulesets/ FS-->>-Core: Configuration + rule definitions
Core->>+Git: Check ADR history, phase state Git-->>-Core: Phase & commit data
Core->>Core: Evaluate rules (Native + OPA) Core-->>-Gateway: ValidationResult { status, issues }
Gateway->>Gateway: 5. Wrap in SuccessEnvelope { success, data, meta } Gateway->>Gateway: 6. Audit log + completion duration
Gateway-->>-Agent: { content: [{ type: "text", text: "{...}" }] }
Note over Agent,Gateway: All errors wrapped in ErrorEnvelope with EvolithErrorCode
| Transport | Use | Command |
|---|---|---|
| stdio (JSON-RPC 2.0) | Local agents, Cursor, Claude Desktop | evolith-mcp serve |
| Streamable HTTP (official MCP SDK) | Remote agents, scalability | evolith-mcp serve --transport http --port 49100 |
The default port is `3000` (
main.ts: envPORTor--port, falling back to3000). The49100used in the examples is an arbitrary value, not the default.
Logs are always written to stderr (Pino), because stdout is reserved for the JSON-RPC stream of the stdio transport.
# From the monorepo
npm install @beyondnet/evolith-mcp
npm install -g @beyondnet/evolith-mcp
# stdio (default) — for Cursor, Claude Desktop, etc.
evolith-mcp serve
evolith-mcp serve --transport http --port 49100
| Variable | Default | Description |
|---|---|---|
TRANSPORT | stdio | Active transport: stdio or http |
PORT | 3000 | Port for the HTTP transport |
MCP_HTTP_HOST | 0.0.0.0 | Bind host of the HTTP server. Use 127.0.0.1 for local-only |
EVOLITH_API_KEY | — | API key for authentication on the HTTP transport |
EVOLITH_MCP_ALLOW_NO_AUTH | false | HTTP only. Allows HTTP to start without an API key (non-production only). Ignored in production and on stdio (which warns at startup) |
JWT_SECRET | — | Optional secret used to validate a Bearer JWT (HS256) in addition to the API key |
NODE_ENV | development | In production, HTTP auth is mandatory |
LOG_LEVEL | info | Pino log level: trace, debug, info, warn, error |
REDIS_URL | — | Redis URL for the resource cache (e.g. redis://localhost:6379). The cache is optional. |
OTEL_EXPORTER_OTLP_ENDPOINT | — | OpenTelemetry endpoint for tracing |
OTEL_SERVICE_NAME | evolith-mcp | Service name used in the traces |
EVOLITH_MCP_REQUEST_STATE_SECRET | — | Key that seals the MRTR requestState (2026-07-28 path). Falls back to EVOLITH_API_KEY, then JWT_SECRET, then a per-process key. Set it explicitly whenever more than one replica is served, or an approval retry landing on another replica is rejected |
EVOLITH_MCP_RESOURCE_AUTH_SERVERS | — | Comma-separated authorization server issuers published in the Protected Resource Metadata document. Defaults to EVOLITH_MCP_OAUTH_ISSUER |
EVOLITH_MCP_RESOURCE_URI | request host | Canonical URI of this server (RFC 8707 resource identifier) published as resource |
EVOLITH_MCP_RESOURCE_SCOPES | read write | Scopes published as scopes_supported and challenged in WWW-Authenticate |
The binary also accepts the flags
--transport/-t,--port/-p,--api-keyand--allow-no-auth(HTTP only), as well as theevolith-mcp versionsubcommand.
No request authentication is required: the process is local, single-user, and the agent runs it directly. The transport establishes an explicit local session principal (id=local-stdio-session, role=local-session, roles=[local-session, operator], scopes=[read, write]) that is recorded in the audit trail of every call (GT-572).
This is not an authorization bypass: ABAC (native + OPA) is still evaluated on every tools/call with that identity, the deploy tools are still denied in production (they require architect), and every mutative tool still demands the HITL gate { apply, approvalToken }. --allow-no-auth / EVOLITH_MCP_ALLOW_NO_AUTH do not apply to stdio (there is no request authentication to skip); if they are passed together with --transport stdio, the server warns about it on stderr at startup.
In production (NODE_ENV=production), authentication is mandatory: validateAuth() ignores EVOLITH_MCP_ALLOW_NO_AUTH and rejects every request without a valid credential (401). The value of EVOLITH_API_KEY is an arbitrary secret (any string; no prefix required) compared by equality. It is accepted in either of these two headers:
Authorization: Bearer <EVOLITH_API_KEY>
x-api-key:
/health is public (a liveness probe) and requires no credential.
When an OAuth issuer is configured, /.well-known/oauth-protected-resource (and the path-inserted /.well-known/oauth-protected-resource/ ) is also public: it is the RFC 9728 document an unauthenticated client reads to discover which authorization server to go to, so gating it behind the credential it is trying to obtain would make discovery impossible. It carries no MCP data — only the issuer, the resource identifier and the scope names. A 401 from a protected resource additionally carries a WWW-Authenticate: Bearer resource_metadata="…", scope="…" challenge, which is the discovery mechanism MCP clients must prefer. Without an issuer the endpoint returns 404 rather than publishing a document with an empty authorization_servers.
Client registration never flows through this server: it is a resource server, and the 2026-07-28 revision has clients obtain a client_id from a Client ID Metadata Document (or pre-registration) at the authorization server. Dynamic Client Registration is deprecated and is not implemented here; protected-resource-metadata.spec.ts fails the build if a registration endpoint is ever introduced.
The server answers two protocol revisions on the same HTTP endpoint:
| Revision | How a request selects it | Shape |
|---|---|---|
2026-07-28 (current) | _meta["io.modelcontextprotocol/protocolVersion"] on every request, or a server/discover call | Stateless. No initialize, no notifications/initialized, no Mcp-Session-Id. Every result carries resultType; the approval gate is expressed as an InputRequiredResult with a sealed requestState (MRTR) |
2025-11-25 | an initialize request | The handshake-based path served by @modelcontextprotocol/sdk, which mints and requires Mcp-Session-Id |
Both revisions run through one dispatch, so ABAC (native + OPA), the scope gate, the approval gate and the audit trail are the same code on either path. The 2025-11-25 path is retained because the published SDK still declares it as its latest revision; it is not an alternative design.
On the 2026-07-28 path a state-changing tool answers resultType: "input_required" with an elicitation/create request and an opaque requestState. The client gathers the human's approval and retries the original call — new JSON-RPC id, same parameters, plus requestState and inputResponses. The requestState is sealed with AES-256-GCM and bound to the principal, the tenant, the originating call and a short TTL, so it cannot be replayed across users, calls or time. A caller that already holds an approval may still pass { apply: true, approvalToken } inline and skip the round trip, exactly as on the 2025-11-25 path.
The
ApiKeyProvisioningService(below) is an advanced and optional mechanism for issuing keys with anevk_prefix, a SHA-256 hash and a TTL. It is independent of the startupEVOLITH_API_KEYdescribed here.
The ApiKeyProvisioningService manages the lifecycle of the keys:
| Operation | Description |
|---|---|
generateKey(label, options) | Generates a key with an evk_ prefix, a SHA-256 hash and a configurable TTL (90 days by default) |
validateKey(rawKey) | Validates the key against the stored hash and checks expiry |
rotateKey(keyId) | Revokes the current key and generates a new one for the same client |
revokeKey(keyId) | Revokes a key immediately |
Keys carry scopes: read, write, admin. They are bound to a tenant.
The AbacEvaluator controls which tools each user may invoke, based on their roles:
| Tool type | Allowed roles | Environment |
|---|---|---|
| Read (list, get, status) | Every authenticated role | Any |
| Write (fix, install, set) | operator, sre, architect, admin | Any |
| Write | developer, qa | Non-production only |
| Deploy (deploy, publish, merge) | architect, admin, operator, sre | Any |
| Deploy | Anyone except architect | Blocked in production |
ABAC codes:
| Code | Cause |
|---|---|
ABAC-01 | Tool denied for the user's role/environment |
ABAC-02 | User with no roles — every tool is denied |
ABAC-03 | Tool not classified into any known group |
Tool classification (substring heuristic). The internal role sets are DEVELOPER = {developer, qa}, OPERATOR = {operator, sre} and ARCHITECT = {architect, admin}. The read/write/deploy classification of each tool is a heuristic over its name (abac-evaluator.ts): it counts as read if the name contains read/list/get (or does not begin with evolith-); as write if it contains write/replace/run/fix/advance; as deploy if it contains deploy/publish/merge. Because of that heuristic, evolith-phase-advance is classified as write (the advance substring), even though it only proposes the transition. A tool that fits no group is rejected with ABAC-03.
Authentication precedence (HTTP). The guard (mcp-server-auth.ts) evaluates the API key first: if the Authorization: Bearer or the x-api-key heade
Only the beginning is shown; read the rest in the repo