
Timezest Mcp
io.github.sagarkalra-techPublished by GitHub account sagarkalra-techv1.1.0Updated Oct 24, 2026
MCP Server for TimeZest: manage appointments, pending requests, and ticket schedules in Claude.
Overview
A read-only MCP server that brings TimeZest scheduling data into an assistant: appointments, pending requests, engineer schedules and ticket-linked bookings.
- What you can do
- Look up appointments by date range, engineer, status or ConnectWise ticket number, and get a morning briefing of today's confirmed bookings grouped by engineer with pending unbooked requests alongside them. Review pending and cancelled scheduling requests, including how long they have been waiting and rebooking links for cancellations. Check one engineer's upcoming schedule and list the appointment types configured in TimeZest. Produce aggregate scheduling statistics such as counts by status and engineer, today's load and the oldest pending request, with timezone-aware display strings.
- When to use it
- Useful for teams running TimeZest with ConnectWise that want to query bookings, chase unbooked requests and review team scheduling load in natural language instead of switching between tabs.
- Requirements
- Runs locally over stdio via the npm package timezest-mcp, so Node.js 20 or newer is needed. A TimeZest API key is required in the TIMEZEST_API_KEY environment variable; optional variables set the default IANA timezone and the look-back and look-ahead windows. Network access to the TimeZest API is required.
Installation
In SourceWeft
- Open Timezest Mcp in the dashboard and add it to a workspace.
- Enable the server for the chats that should use its tools.
Desktop only via STDIO. STDIO servers start a local process, so they need the SourceWeft desktop host.
Other MCP clients
Follow the launch instructions in the repository.
README
⏰ TimeZest MCP Server
[npm version] [MCP Registry] [CI] [License: MIT]
Bring your TimeZest scheduling data directly into Claude through the Model Context Protocol (MCP). Perform real-time appointment lookups, engineer briefings, and ticket-linked schedule management — all with natural language.
Registry ID: io.github.sagarkalra-tech/timezest-mcp
🌟 Why TimeZest MCP?
If you manage a team that uses TimeZest with ConnectWise, you know the pain: toggling between tabs, manually checking who's booked, chasing unbooked scheduling requests, and cross-referencing ticket numbers.
This MCP server turns Claude into a scheduling assistant that can:
- 🗓️ Give you a morning briefing of today's confirmed appointments grouped by engineer
- 🔍 Find appointments by ticket number — just say "find appointments for ticket #964400"
- ⏳ Surface aging unbooked requests that need follow-up
- 📊 Generate aggregate scheduling stats across your team
- 🕐 Handle timezone-accurate reporting across global MSP teams
🚀 Get Started in 30 Seconds
Prerequisites
- Node.js 20+ — Download here
- TimeZest API Key — Found in TimeZest → Settings → API
Option 1: Claude Desktop (Recommended)
Add this to your Claude Desktop configuration file:
- Windows:
%APPDATA%\Claude\claude_desktop_config.json - macOS:
~/Library/Application Support/Claude/claude_desktop_config.json
Restart Claude Desktop. You're done.
Option 2: Claude Code CLI
Option 3: VS Code (Copilot)
Add to your .vscode/settings.json or User Settings:
🛠️ Available Tools
The server exposes 8 specialized tools that Claude can call:
💬 Example Prompts
Once connected, just talk to Claude naturally:
⚙️ Configuration
🏗️ Architecture
Key Design Decisions
- Read-Only: Intentionally kept read-only — no mutations to your scheduling data
- Resilient Client: Exponential backoff with automatic
Retry-Afterheader support for 429/5xx errors - Deep Pagination: Recursively crawls all pages (50-item chunks) to build complete datasets
- Timezone Precision: Converts Unix timestamps to strict ISO with localized display strings using IANA timezone identifiers
- Engineer Resolution: Uses a "highest-confidence" search — checks
scheduled_agentsfirst, then falls back toresources - Warm Cache: Appointment types are cached after first fetch to reduce API overhead
🧪 Testing
The server ships with a comprehensive Vitest test suite — 22 tests across 3 files:
All tests use mocked API responses — no live API calls, no secrets needed.
🔒 Security & Trust
- NPM Provenance: Every release is cryptographically signed via GitHub Actions OIDC
- CI/CD Pipeline: Automated build → test → publish on tagged releases (no manual
npm publish) - No Secrets in CI: Tests require no API keys — fully deterministic with mocked responses
- Node.js Matrix: Tested on Node 20 (LTS) and 22 (Current)
📦 Distribution
🤝 Contributing
- Fork the repository
- Create a feature branch (
git checkout -b feat/my-feature) - Make your changes and ensure tests pass (
npm test) - Submit a pull request
See ARCHITECTURE.md for a deep dive into the data flow, component design, and transformation logic.
📄 License
MIT © 2026 Sagar Kalra
Source: README.md at commit 766cd8b
Tools
0Version history
1- v1.1.0LatestSep 16, 2026
Score
40% of the applicable checks met
How complete this listing is, computed from the registry entry, the endpoint and public sources. Not a rating of the server and not a statement that it is safe or working today: reachability is a dated check, and no installs or reviews go into the score.
- Reached in the last 30 days25 pts
Keep the endpoint answering the MCP handshake; SourceWeft probes it every 30 days.
- Auth declared10 pts
Declare the credential headers in the registry entry, or keep the endpoint answering the probe so its auth can be classified.
- Tools listed15 pts
Answer tools/list without a credential, or ship a package that starts in a sandbox without one.
- DescriptionRequired15 pts
- Logo10 pts
- Documentation15 pts
- License10 pts
Declare a license in the repository or the registry entry.