Pet Log MCP Server
This server does not currently have a deployable release on Glama. Follow the repository's setup instructions to install and configure it for your MCP client, including any required credentials and local dependencies.
Pet Log MCP Server
A local MCP server for household pet care notes: vet-visit histories, medication organizers, weight trends, feeding notes, and on-demand groomer/sitter handoff summaries. Plain Python, standard library only, JSON-RPC over stdio — exactly what an MCP client expects. No network calls, ever.
Education and organizing only. This tool is a notebook, not a vet. It stores and summarizes your own notes; it never gives medical, medication dosing, or feeding advice. Always consult your veterinarian for medical decisions.
Made by Asdale Dynamics, LLC — plain-English AI education and practical local-first tools. Licensed MIT.
Why this exists
The companion project new-pet-first-week-guide is the setup manual for a pet's
first days home. This project is the ongoing log: the place where the
checkup records, medication schedules the vet gave you (organized by you,
corrected by you), weigh-ins from your own scale, food notes, and hand-off
notes for the groomer or pet sitter live for the long haul.
Related MCP server: notes-box
How it works
One JSON folder you own. All data lives in six plain-JSON files (
pets.json,vet_visits.json,medications.json,weights.json,feeding_notes.json,handoff_notes.json) in a single folder. Open it, read it, back it up, keep it in your own cloud drive if you like.MCP-style stdio server. Speaks JSON-RPC 2.0 (the same shape the Model Context Protocol uses) with
initialize,notifications/initialized,ping,tools/list, andtools/call. One JSON message per line on stdin, one response per line on stdout; log messages go to stderr only.Zero dependencies at runtime. Python 3.10+ standard library only.
pytestis a test-time dependency only.
Tools (13)
Tool | Purpose |
| Register and list the pets in the log. |
| Build a per-pet vet-visit history. |
| Organize the schedules your vet gave you (never dosing advice). |
| Record weigh-ins and see the trend with per-entry change. |
| Log feeding observations, newest first. |
| Groomer/sitter hand-off summaries, assembled on demand. |
| Remove one wrong or duplicated record by id. |
Quick start
Run the server directly (no install needed, from the repo folder):
python -m pet_log_mcpName the data folder explicitly (default is ./pet-log-data):
PET_LOG_DATA_DIR=/home/you/pet-logs python -m pet_log_mcpOr install it with pip to get the pet-log-mcp command:
pip install .Manual smoke test
printf '%s\n' \
'{"jsonrpc":"2.0","id":1,"method":"initialize","params":{}}' \
'{"jsonrpc":"2.0","id":2,"method":"tools/list"}' \
'{"jsonrpc":"2.0","id":3,"method":"tools/call","params":{"name":"add_pet","arguments":{"name":"Mochi","species":"cat"}}}' \
| python -m pet_log_mcpWiring it into an MCP client
Point your MCP client at the process python -m pet_log_mcp (transport:
stdio, no args). If the client app stores mcpServers config, a local-first
example looks like:
{
"mcpServers": {
"pet-log": {
"command": "python",
"args": ["-m", "pet_log_mcp"]
}
}
}Set PET_LOG_DATA_DIR in the environment (or add an "env" block, if your
client supports it) to keep the JSON folder wherever you keep personal data.
Tests
python -m pytestThe test suite covers the store, every tool (through mocked/fake persistence), the full JSON-RPC protocol surface, and a subprocess end-to-end run of the real stdio server. No test touches the network.
Privacy, first and always
Everything stays in the local JSON folder. Nothing is sent anywhere.
No telemetry, no analytics, no embedded API keys, no commitments.
Because it's your data: adding a pet's name in the log is on you; the file format is plain JSON so you can edit or export anytime.
Disclaimer
Pet Log MCP Server is an organizing and education tool. It does not provide veterinary diagnosis, treatment, dosing, or feeding recommendations and must not be used as a substitute for professional veterinary care.
License
MIT — see LICENSE. Copyright (c) 2026 Asdale Dynamics, LLC.
This server cannot be deployed
Maintenance
Related MCP Connectors
Agent-native notes, tasks, dev-docs, vaults, sync & handoffs. MCP + OpenAPI dual surface.
Persistent memory layer for AI tools. Save and recall notes across Claude and other MCP clients.
Local-first task manager: create, edit, and complete tasks, projects, and checklists via MCP.
MCP-native notes and memory for ChatGPT, Claude, and other AI tools.
Related MCP Servers
- AlicenseNot gradedqualityCmaintenanceEnables managing a local JSON notes file through five MCP tools: add, get, update, delete, and list notes, with atomic saves and a write lock.MIT
- FlicenseNot gradedqualityCmaintenanceEnables MCP clients to add, retrieve, update, delete, and list notes stored in a configurable local JSON file, with atomic saves behind a write lock.-
- AlicenseNot gradedqualityCmaintenanceEnables any MCP-capable AI app to read and write a single, locally stored, user-owned journal file, so memory persists across different assistants without a cloud account. Entries are append-only and cryptographically chained, with tools for waking briefings, writing notes, searching, pinning, verifying integrity, and exporting backups.AGPL 3.0
- AlicenseNot gradedqualityAmaintenanceEnables reading, searching, and safely editing Apple Stickies on macOS through a local MCP client, with version-checked writes, backups, undo, and recoverable delete/restore. Runs entirely on-device over stdio, using Accessibility for live-note edits while refusing to overwrite data on conflicts or missing permissions.1MIT