Skip to content
MCP serverOpen-source agent

Dayfile.app

Append-only personal event log served as JSON Feed, Atom and an MCP server for assistants

Dayfile.app is an open-source MCP server for productivity. It is available on Web, CLI, API and Self-hosted.

Pricing
Open source
Platforms
Web, CLI, API
Listing updated

Similar MCP servers

About Dayfile.app

The README describes dayfile as a personal, append-only event log that is published as feeds (JSON Feed, Atom and CloudEvents). Anything that can send an HTTP request can write to it, including scripts, webhooks, devices and AI assistants connected over MCP. Changes to an event are recorded as new events that point back to the original, so a to-do's status is derived from that chain rather than by editing rows.

Feeds are addressed by URL queries such as tag and status filters, and named feeds can be saved as private or public. Access is controlled per credential: read tokens can be sent as a Bearer header or in the path for feed readers that cannot set headers, and write-only tokens let remote clients add events. A JSON API adds long-poll and live SSE streams alongside the feeds.

The project is aimed at people who want a self-hostable, scriptable log of their own events and to-dos, and who are comfortable with the command line. The README offers a hosted option at dayfile.app with a free beta, and a self-hosted option installed with pip or run with Docker or systemd.

Tools this server provides (5)

Taken from the project documentation; the running server may differ.

dayfile_list_events
List events in the log, filtered by query.
dayfile_get_event
Fetch a single event.
dayfile_add_event
Append a new event to the log.
dayfile_close_todo
Close an open to-do by appending a state-change event.
dayfile_retract_event
Retract an event by appending a retraction event.

What it does

Append-only event log
Events are never edited; state changes are new events that reference the original, enforced by SQLite triggers.
Feeds and filtered views
Serves JSON Feed, Atom and CloudEvents, with query-based filters and saved named feeds.
Per-credential access
Read and write tokens decide which audiences a client can see or write to, and tokens can be revoked.
MCP server for assistants
Exposes list, get, add, close and retract tools so an assistant such as Claude can read and write the log.
Heartbeat monitoring
Sources declare how often they should report, and an open to-do is written when a source goes quiet.
Self-hosting options
Runs as a pip-installed CLI and server, or via Docker or systemd, with optional adapters for email and push alerts.

Good for

  • Keep a personal to-do list that can be read in a feed reader on any device
  • Let an AI assistant add and close tasks in a personal log over MCP
  • Collect events from scripts, webhooks or cron jobs into one searchable log
  • Run a private, self-hosted event feed on your own server

Pricing

Open source

Free beta on dayfile.app; self-hosting under AGPL-3.0. Billing is not built.

Dayfile.app FAQ

Is dayfile open source?

The README states it is free software under the GNU Affero General Public License v3 (AGPL-3.0-only).

Can I use Dayfile.app without running a server?

Yes. The README describes a hosted option at dayfile.app, where a log is available at a username subdomain with a free beta, and the MCP endpoint is provided there.

How do I connect Dayfile.app to an AI assistant?

The README says that in Claude you can add a custom connector with the log's MCP address and sign in with GitHub. Locally, the dayfile mcp command runs over stdio with read and write tokens set in environment variables.

Is billing available?

The README states that billing is not built.

Is Dayfile.app free?

Yes. Dayfile.app is open source, so you can use and self-host it for free.

What are the best alternatives to Dayfile.app?

The closest alternatives to Dayfile.app are Worklog, To-do Calendar, Devlog, Todo MCP and Monologue.

Where this comes from

  • Description: written from the project's README.

How AIPoweredBot checks servers