Spec Driven Development
MCP Server that facilitates spec-driven development workflows, not just Vibe Coding.
Description from the maker (original English)
Read-only MCP server for official Tauri 2 documentation snapshots.
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": {
"tauri-docs": {
"command": "npx",
"args": [
"-y",
"tauri-docs-mcp"
]
}
}
}Run this in your terminal
claude mcp add tauri-docs -- npx -y tauri-docs-mcpPaste into .cursor/mcp.json in your project, or ~/.cursor/mcp.json
{
"mcpServers": {
"tauri-docs": {
"command": "npx",
"args": [
"-y",
"tauri-docs-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": {
"tauri-docs": {
"type": "stdio",
"command": "npx",
"args": [
"-y",
"tauri-docs-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 fxckcode, 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 (0.2.0) reached the registry 2 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, FOGA node
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.
A Finch OG Agent cell as an MCP server: read the world, dry-run and take ERC-8414 tenders by policy.
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.
A read-only Model Context Protocol (MCP) server for resolving and querying a checked-in snapshot of official Tauri 2 documentation.
The committed package-lock.json is authoritative. Use npm install for local development and dependency changes; CI and reproducible installs use npm ci. Lockfile changes must be committed with dependency changes.
Install the published package:
npm install --global tauri-docs-mcp
Or install it in a project and invoke the binary from node_modules/.bin:
npm install tauri-docs-mcp
npx tauri-docs-mcp
Build from a checkout, then run the compiled stdio server:
npm install
npm run build node dist/src/index.js
Configure an MCP client to launch the installed binary over stdio:
{
"mcpServers": { "tauri-docs": { "command": "tauri-docs-mcp" } } }
Portable configurations for generic MCP clients, Claude Desktop, Cursor, and VS Code are in examples/mcp/. They pin the package version and use stdio only. Installable agent guidance is in skills/tauri-docs-mcp/SKILL.md.
The server writes MCP JSON-RPC messages only to stdout. It does not start an HTTP listener. Diagnostics are disabled by default; opt in with TAURI_DOCS_MCP_DIAGNOSTICS=1 and they are written to stderr as event names only (never request arguments, corpus content, secrets, or PII).
The server exposes two stable tools and one backwards-compatible legacy tool.
resolve_tauri_docsResolve the default snapshot or an explicit identifier:
{}
{ "identifier": "tauri@2" }
Supported identifiers are tauri, tauri@2, and the immutable snapshot ID tauri-2@58194ceb69424c4332b2780b196ced3a6fffb32b. Resolution is deterministic and returns snapshot, source revision, generated-at, version, and entry-count metadata. Unsupported identifiers return UNSUPPORTED_SNAPSHOT.
query_tauri_docsQuery content from the default or selected snapshot:
{ "query": "capabilities", "snapshot": "tauri@2", "limit": 3 }
query is required and must contain 1–200 non-whitespace characters and at most 16 whitespace-separated terms. snapshot is optional and accepts the resolver identifiers. limit is an optional integer from 1–10 (default 5). Results are bounded, deterministic, and include title, heading, section, canonical URL, content, corpus snapshot, source revision, Tauri version, and versionSensitive. Empty matches return results: [], not an application error. Inputs reject unknown fields.
Both tools are read-only and idempotent. They read only the checked-in corpus; there is no live fetching or arbitrary URL access.
The shared tool contract enforces a 200-character query, 16 terms, 10 results, 1,200-character snippets, 8 KiB corpus entries, and 32 KiB serialized tool responses. The stdio frame buffer is capped at 64 KiB. EOF, transport errors, SIGINT, and SIGTERM share an idempotent shutdown path with a one-second cleanup bound; in-flight work is allowed to settle or is abandoned at that bound.
search_tauri_docs (compatibility)Existing clients can continue to call:
{ "query": "capabilities", "limit": 3 }
It remains supported with its existing response shape and limits. New clients should use resolve_tauri_docs followed by query_tauri_docs.
Application errors are returned with stable discriminated codes such as INVALID_INPUT, INVALID_QUERY, INVALID_LIMIT, and UNSUPPORTED_SNAPSHOT; protocol errors remain MCP protocol errors.
This package is intentionally:
https://v2.tauri.app/.It never fetches user-controlled URLs, writes files, executes arbitrary code, controls a Tauri runtime, provides hosted HTTP, authenticates users, or emits telemetry. See docs/architecture.md for the module seams.
npm install # install from package-lock.json
npm run typecheck # TypeScript validation npm run build # compile runtime files to dist/ npm test # run the full Vitest suite npm run lint # ESLint npm run format # format tracked source and documentation npm run format:check # verify formatting without writing npm pack --dry-run # inspect the publish allowlist npm run package:smoke -- /path/to/tauri-docs-mcp-0.1.0.tgz # test a packed install
The pull-request workflow runs the same commands from a clean checkout. To reproduce its package step locally, create a tarball and pass its path to the smoke test:
mkdir -p .tmp/package
TARBALL=$(npm pack --json --pack-destination .tmp/package | node -e "let s=''; process.stdin.on('data', d => s += d).on('end', () => process.stdout.write(JSON.parse(s)[0].filename))") npm run package:smoke -- ".tmp/package/$TARBALL" rm -rf .tmp
See CONTRIBUTING.md for contribution rules and SUPPORT.md for supported use cases. Report vulnerabilities privately as described in SECURITY.md.
The release workflow runs from a version tag and checks package, immutable corpus metadata, registry metadata, and the pack allowlist before publishing. npm publishing uses provenance/trusted publishing where enabled, then runs a fresh published-package smoke gate. Official MCP Registry publication is a separately protected environment chained after the successful npm job. No release is published from pull requests. See CHANGELOG.md for the release and MCP contract-change process.
The runtime reads only the checked-in generated snapshot under corpus/; it never fetches documentation URLs. The approved manifest records canonical URLs, section hierarchy, Tauri version, source revision/freshness metadata, and a SHA-256 hash for each concise excerpt. The source revision is immutable, so a future snapshot must use a new tauri-2@ directory.
After an approved source review, regenerate and validate deterministically:
npm run ingest:docs
npm run validate:corpus
validate:corpus fails if the manifest is unsafe or incomplete, if content hashes drift, or if the generated artifact differs from its manifest. Source and license attribution is preserved in corpus/NOTICES.md.