timex-mcp
Enables coding agents to create, start, pause, and delete labeled timers within projects, with persistence via SQLite, through a stdio MCP interface.
README
Timex
Deck-styled native scheduler for project-scoped, time-boxed work, exposed to coding agents over MCP.
Layout
app/is the Native SDK app package. It contains the UI, model, assets, and Native manifest.mcp/is the standalone Zig package for SQLite storage,timex-mcp, and thetimex-storehelper CLI.- The root
build.zigonly orchestrates both packages. It does not import or require a checked-in Native SDK dependency.
Features
- Create and select projects.
- Create timers with a required duration, optional details, and an optional RFC 3339 calendar start.
- Start, pause, reset, and delete timers; running timers count down and continue into overdue time.
- Automatically start the oldest due timer while the Native app is open, while preserving schedule drift when another timer is active.
- Deliver one-time macOS notifications for due scheduled work while Timex is open.
- Persist projects and timers in SQLite.
- Run
timex-mcpas a stdio MCP server for agent control. - Let the Native app talk to storage through
timex-store, so SQLite is isolated from the app binary.
Build
zig build # builds timex-mcp, timex-store, and the Native app
zig build mcp # builds only the standalone MCP/store binaries
zig build app # builds only the Native app via `native build app`
zig build run # runs the Native app with the MCP server sidecar
zig build test # runs MCP/store tests and Native app tests
The root MCP binaries install to zig-out/bin/timex-mcp and zig-out/bin/timex-store.
Native app artifacts are produced by the Native CLI under app/zig-out/.
Runtime
- Set
TIMEX_DB_PATH=/path/to/timex.sqliteto use a specific SQLite database. - Set
TIMEX_STORE_BIN=/path/to/timex-storewhen launching the app from a working directory wherezig-out/bin/timex-storeis not available. - MCP clients should launch
zig-out/bin/timex-mcp.
Use Timex with a coding agent
Timex is a local stdio MCP server. Build it before starting an agent:
zig build mcp
The repository includes a project-local Codex configuration at .codex/config.toml:
[mcp_servers.timex]
command = "zig-out/bin/timex-mcp"
startup_timeout_sec = 10
tool_timeout_sec = 30
Start Codex from the repository root so the relative command resolves to the freshly built binary:
codex
Codex starts the MCP server when the session opens. After rebuilding timex-mcp, restart Codex so the session uses the new binary; an already-running MCP process does not hot-reload.
You can verify the server from Codex with a prompt such as:
Use Timex to create a project named
Codex test, create a 25-minute timer namedMCP smoke testin it, start the timer, and report the resulting IDs and state.
For a disposable test database, set TIMEX_DB_PATH before launching Codex:
TIMEX_DB_PATH=/tmp/timex-codex-test.sqlite codex
If Timex tools do not appear, confirm that zig-out/bin/timex-mcp exists and that Codex was started from this repository root so .codex/config.toml and its relative command are resolved. If the tools appear but calls fail after a rebuild, restart Codex before debugging the installation—the current session may still be connected to the previous server process.
To add Timex to another coding agent, use its stdio MCP configuration and point command at the built binary. The command must run with the repository root as its working directory, or use an absolute path:
{
"mcpServers": {
"timex": {
"command": "/absolute/path/to/timex/zig-out/bin/timex-mcp",
"env": {
"TIMEX_DB_PATH": "/absolute/path/to/timex.sqlite"
}
}
}
}
Use the equivalent mcpServers.timex entry for agents whose configuration is TOML or YAML. Keep the server on stdio; it does not expose an HTTP endpoint. The available tools and resources are documented in skills/timex-agent/SKILL.md.
Recommended Servers
playwright-mcp
A Model Context Protocol server that enables LLMs to interact with web pages through structured accessibility snapshots without requiring vision models or screenshots.
Magic Component Platform (MCP)
An AI-powered tool that generates modern UI components from natural language descriptions, integrating with popular IDEs to streamline UI development workflow.
Audiense Insights MCP Server
Enables interaction with Audiense Insights accounts via the Model Context Protocol, facilitating the extraction and analysis of marketing insights and audience data including demographics, behavior, and influencer engagement.
VeyraX MCP
Single MCP tool to connect all your favorite tools: Gmail, Calendar and 40 more.
graphlit-mcp-server
The Model Context Protocol (MCP) Server enables integration between MCP clients and the Graphlit service. Ingest anything from Slack to Gmail to podcast feeds, in addition to web crawling, into a Graphlit project - and then retrieve relevant contents from the MCP client.
Kagi MCP Server
An MCP server that integrates Kagi search capabilities with Claude AI, enabling Claude to perform real-time web searches when answering questions that require up-to-date information.
Neon Database
MCP server for interacting with Neon Management API and databases
E2B
Using MCP to run code via e2b.
Exa Search
A Model Context Protocol (MCP) server lets AI assistants like Claude use the Exa AI Search API for web searches. This setup allows AI models to get real-time web information in a safe and controlled way.
Qdrant Server
This repository is an example of how to create a MCP server for Qdrant, a vector search engine.