AirysChat MCP Server

Name: AirysChat MCP Server
Author: airys.chat

Remote

airys.chatOther0Updated today

Repository Visit website

About AirysChat MCP Server

AirysChat MCP Server connects AI agents and MCP clients to AirysChat operations, including conversations, CRM, AI agents, WhatsApp campaigns, and administration. It also includes client plugins and shared skills for defined operational workflows.

Key features

01Manage conversations, contacts, and CRM
02Operate AI agents and Meta Business Agent
03Manage WhatsApp templates, campaigns, and flows
04Access analytics, administration, and Help Center
05Use plugins and shared operational skills

Use cases

01Review conversations and CRM records
02Manage WhatsApp campaigns and flows
03Supervise AI agents and review analytics
04Audit accounts and support operations

MCP support

The server exposes AirysChat operations through MCP tools, including an embedded API catalog, and provides three interactive MCP Apps. It supports stdio and stateless Streamable HTTP.

Configuration

Remote servers need no local install. Pick your client, then copy the config or install in one click.

Endpointhttps://mcp.airys.chat/mcp

Copy install command

Claude CodeTerminalclaude mcp add --transport http chat-airys https://mcp.airys.chat/mcp

Cursor.cursor/mcp.json{"mcpServers":{"chat-airys":{"type":"http","url":"https://mcp.airys.chat/mcp"}}}Open in Cursor

VS Code.vscode/mcp.json{"servers":{"chat-airys":{"type":"http","url":"https://mcp.airys.chat/mcp"}}}Open in VS Code

Claude Desktopclaude_desktop_config.json{"mcpServers":{"chat-airys":{"type":"http","url":"https://mcp.airys.chat/mcp"}}}

Windsurf~/.codeium/windsurf/mcp_config.json{"mcpServers":{"chat-airys":{"serverUrl":"https://mcp.airys.chat/mcp"}}}

Run commands in a terminal; paste JSON into the client’s MCP config and restart. Servers that need sign-in will prompt on first connect.

Packages

npm@airyschat/mcp1.0.1stdio

Tools

Details

From the project README

AirysChat MCP

AirysChat for AI agents. This monorepo ships:

@airyschat/mcp — an MCP server that lets Claude, ChatGPT, Codex, Gemini, Antigravity, Muse and any MCP client work with AirysChat: conversations, contacts, CRM, AI Agents, Meta Business Agent, WhatsApp templates/campaigns/flows, analytics reports, account administration, Help Center, and the Airys Platform API (api.airys.chat).
Plugins for 4 agent clients: Claude Code, OpenAI Codex, Google Gemini CLI, and Google Antigravity.
Skills (skills/*) that encode safe AirysChat workflows (airyschat, triage, campaign-check, campaign-retro, crm-pipeline-review, agent-supervision, account-audit, number-health-watch).

Status — 212 curated & catalog tools across 9 toolsets (80 stable, 98 beta, 34 experimental) + embedded 807-endpoint catalog (airyschat_api_search + airyschat_api) + 3 interactive MCP Apps (ui://airyschat/*). See the Quickstart Guide, Generated Tool Reference, Safety Model, and Release Train.

Why this exists

Every major assistant now speaks MCP and installs plugins/skills. AirysChat customers increasingly operate through those assistants instead of a dashboard. The goal is simple: every AirysChat action reachable by any agent, safely — from a local install or a hosted URL (mcp.airys.chat).

Roadmap at a glance

Release	Theme	Target
0.2.0	Foundation + API catalog (`airyschat_api_search`)	2026-11-09
0.3.0	CRM & contacts	2026-11-30
0.4.0	AI agents & Meta Business Agent	2026-12-14
0.5.0	WhatsApp complete & reports	2027-01-11
0.6.0	Hosted MCP on Cloud Run + scoped MCP tokens	2027-03-01
0.7.0	MCP Apps, elicitation, tasks	2027-03-22
0.8.0	Admin & help center	2027-04-12
0.9.0	Airys platform API toolset	after platform M2
1.0.0	GA: public npm, MCP Registry, 4 client plugins, connector directories	2027-05-31

Coverage is computed from the AirysChat routes: docs/api/coverage.md.

Repository

Path
`packages/server`	The MCP server (`stdio` + stateless Streamable HTTP)
`plugins/`	Client plugin bundles (`claude`, `codex`, `gemini`, `antigravity`)
`skills/`	Canonical shared operational skills synced into every plugin bundle
`server.json`	Official MCP Registry manifest (`chat.airys/mcp`)
`specs/`	Lanes, tasks and verification — the plan
`docs/tools/reference.md`	Generated reference for all 212 tools
`docs/quickstart.md`	Per-client setup (Claude, Codex, Gemini CLI, Antigravity, Cursor, Muse)
`docs/safety-model.md`	Indirect prompt-injection fencing, write gates, elicitation, SSRF
`docs/stability-policy.md`	API stability policy & frozen `stable` tool surface
`docs/guides/`	Meta WhatsApp Business Tools MCP complementarity & Meta Muse setup
`docs/adr`	Architecture decisions
`docs/api`	API inventory and coverage
`docs/research`	Market research (dated)
`docs/architecture.md`	How the pieces fit
`AGENTS.md`	Guide for coding agents (rules, skills, subagents)

Development

nvm use            # Node 24
corepack enable    # pnpm
pnpm install
pnpm check         # typecheck + smoke + inventory drift

See CONTRIBUTING.md for the feature/release workflow and SECURITY.md for reporting vulnerabilities.

License

MIT

Comments

Markdown supported

New to MCP?

From what the protocol is to installing your first server, these answers cover most first questions.

Read the getting-started guide

An open protocol that defines how AI applications connect to external tools and data in a uniform way. Clients talk to the model; MCP servers provide tools, resources and prompt templates. They exchange standard messages, so one server can be reused by many clients.