Melaya

Melaya MCP Server

Let an AI assistant use your Android phone, your browser, and the rest of your Melaya account.

Also listed on mcp.so · mcpserver.dev · mcpmarket · cursor.directory · Glama · Product Hunt · AlternativeTo · SaaSHub · AI Agents Directory · AgentLocker · LaunchKiwi

Documentation · Melaya · Device Control · MCP Server

Melaya pairs a phone to your account and gives an agent the same view of it a person has: it reads the screen through Android's accessibility tree, then taps, types, swipes, and moves between apps. No per-app integration, no vendor API. If you can use the app, so can the agent.

This is a remote Model Context Protocol server. MCP is a vendor-neutral standard, so one endpoint works everywhere:

https://api.melaya.org/mcp

Nothing to install, no SDK required.

Connect

Claude Code

claude mcp add --transport http melaya https://api.melaya.org/mcp

Codex CLI

codex mcp add melaya --url https://api.melaya.org/mcp

Direct HTTP needs experimental_use_rmcp_client = true in ~/.codex/config.toml.

Cursor

~/.cursor/mcp.json, or .cursor/mcp.json for one project:

{ "mcpServers": { "melaya": { "url": "https://api.melaya.org/mcp" } } }

Claude

claude.ai, Desktop and mobile: Settings → Connectors → Add custom connector, paste https://api.melaya.org/mcp

ChatGPT

Settings → Connectors → Developer mode, paste the same endpoint

Le Chat

Connectors → Add custom MCP connector

VS Code

code --add-mcp '{"name":"melaya","type":"http","url":"https://api.melaya.org/mcp"}'

Everything else

Windsurf, Zed, Cline, Goose, Lovable, Gemini CLI, Qwen Code and any other MCP client take the same block, in whichever file that client uses for MCP servers:

{ "mcpServers": { "melaya": { "url": "https://api.melaya.org/mcp" } } }

Authentication is OAuth 2.1 with PKCE. You choose which permissions to grant on a Melaya consent page; the assistant receives a scoped token. Your password is never shared.

Then just ask:

Connect my phone and go through my unread Instagram DMs.

---

Give your AI the Melaya skill

This server gives your assistant the tools. The [Melaya skill](https://github.com/melaya-labs/melaya-mcp/tree/main/skills/melaya) gives it the method: which tool to call first, how to connect services, build and validate a pipeline on a real run, set triggers and approvals, and read results. Install both and your assistant gets real work right the first time instead of guessing.

Any assistant (ChatGPT, Gemini, Cursor, …): paste this line.

Install the Melaya skill from https://github.com/melaya-labs/melaya-mcp/tree/main/skills/melaya and use it whenever I ask you to work with Melaya.

Claude Code: the Melaya plugin from this repo already bundles it, next to melaya-setup. Installing the plugin is enough. To use the skill on its own, copy it into your skills folder, then restart. It loads on its own when you mention Melaya.

git clone --depth 1 https://github.com/melaya-labs/melaya-mcp /tmp/melaya-mcp && cp -r /tmp/melaya-mcp/skills/melaya ~/.claude/skills/

Claude.ai and Claude Desktop: download skills/melaya as a .zip and upload it under Settings → Capabilities → Skills. Anywhere else, add SKILL.md as a project instruction or rules file.

Starts right Opens every session with melaya_setup_status and closes each gap it reports before doing anything else. Stays safe Never asks for a key in chat, never decides an approval for you, confirms before anything is sent or deleted.

Builds pipelines that run Reads the config before changing it, previews before saving, and judges a run by its real output, not its status. Loads only what it needs One short router plus eleven modules: quickstart, pipelines, agentic systems, data, triggers and approvals, devices, handover.

Content creation skill

[`skills/melaya-content-creation`](./skills/melaya-content-creation) is a separate, brand-neutral skill for any AI: launch, promo, explainer and social videos, films, carousels, thumbnails, captions and share copy for any brand or product, from the brief and the brand kit to the render, QA and publishing package. It builds with HyperFrames (Remotion as an alternative) and hands off a ready-to-render package when no renderer is available. One short router plus a graph of modules loaded on demand, style presets as data, templates and worked examples. It does not need the MCP server. In Claude Code the plugin bundles it with the /create-content command; on its own:

git clone --depth 1 https://github.com/melaya-labs/melaya-mcp /tmp/melaya-mcp && cp -r /tmp/melaya-mcp/skills/melaya-content-creation ~/.claude/skills/

It adapts the launch-video method of brag (MIT) and follows the HyperFrames composition contract (Apache-2.0); credits in the skill's README.

Obsidian vault skill

[`skills/melaya-obsidian-vault`](./skills/melaya-obsidian-vault) builds the most complete, well-linked Obsidian vault of your own knowledge, end to end, from everything you have: Drive, Docs, Sheets, Slides, Notion, Confluence, OneNote, Evernote, Gmail, Outlook, Slack, Teams, Jira, GitHub, Zoho, web pages and local files (PDF with OCR, Office, images, transcripts). Solo founder, team wiki, company handbook, research vault, client project, migration, personal knowledge base, multilingual, public plus private zones, ongoing refresh. It starts from a one-page charter, inventories every source with permission and personal-data checks, reads them read-only in resumable batches with a source ledger, and writes typed notes with properties, sources and dates on every fact, hubs, a deliberate link graph (no orphans, three hops from Home), Bases and Canvas views. Stdlib-only Python scripts scaffold the vault, keep the ledger, generate hubs, auto-link the glossary, add outlines, check every rule (links, properties, taxonomy, zones, secrets, personal data), report graph metrics and export the public zone. With the MCP server it reads through Melaya's read-only connectors and can write to a connected vault; without it, it works from exports and plain Markdown. In Claude Code the plugin bundles it with the /build-vault command; on its own:

git clone --depth 1 https://github.com/melaya-labs/melaya-mcp /tmp/melaya-mcp && cp -r /tmp/melaya-mcp/skills/melaya-obsidian-vault ~/.claude/skills/

[`skills/excom-role-skills`](./skills/excom-role-skills) gives AI executives the playbooks of their roles: 125 generic skills in 15 role folders (CEO, CFO, CRO, CMO, COO, CTO, CPO, customer, data, people, security, legal, quality, supply chain, partnerships), one per recurring job, such as the weekly board brief, cash runway, lead qualification, SEO content coverage, incident root cause, contract review or partner onboarding. Each skill first reads your company's own goals, terms and definitions from your knowledge base, names the exact tools and says which steps wait for your approval. They are the role packs Melaya EXCOM ships built in. Several adapt open-source skills under MIT and Apache-2.0, credited per role in THIRD_PARTY.txt (index). On its own:

git clone --depth 1 https://github.com/melaya-labs/melaya-mcp /tmp/melaya-mcp && cp -r /tmp/melaya-mcp/skills/excom-role-skills ~/.claude/skills/

What it can do

Operate your phone

Read the screen, open apps, tap, type, scroll, swipe, screenshot. You watch it work, and one control stops everything.

Melaya ships navigation playbooks for common apps, so the agent arrives knowing where things are instead of exploring blindly.

Operate a browser

The same read-act-verify loop on a desktop site, through the Melaya extension on Chrome or Edge. You attach the tab; the model never picks one. Several conversations, pipelines and the extension panel can work in the same browser at once, each in its own tab (up to 8): attach returns an agent id that the conversation passes on every browser call, and a tab another agent is using is never touched.

<img src="ass