pumaDB vs Obsidian

Memory for agents, notes for humans.

pumaDB and Obsidian both help preserve context, but they are designed for different users. pumaDB gives agents and server-side apps a hosted JSON memory API. Obsidian gives people a local-first Markdown workspace for writing, linking, and organizing ideas.

pumaDB

Hosted JSON memory API

Obsidian

Local-first notes workspace

Quick read

Choose pumaDB when an agent needs memory it can call directly. Choose Obsidian when a person needs a thinking space for private Markdown notes.

Obsidian is where people think in Markdown. pumaDB is where agents remember in JSON.

Best fit

Use the tool that matches the size of the job.

Choose pumaDB for

Hosted MCP memory for agents without vault file access
Structured JSON records such as preferences, project facts, task state, resources, and handoff notes
Server-side REST calls from Workers, API routes, CLIs, scripts, and agent infrastructure
Reviewable memory with row versions, restore, typed safe memory, and explicit cleanup
Small shared memory that multiple tools can query without parsing Markdown files

Choose Obsidian for

Human-authored notes, journals, research logs, and connected writing
Local-first Markdown files that stay useful outside the app
Links, backlinks, tags, graph navigation, canvas boards, and community plugins
Private personal knowledge management on desktop and mobile
Publishing, syncing, and organizing notes through Obsidian's app ecosystem

Detailed comparison

Where they differ.

Area pumaDB Obsidian

Primary job A small hosted JSON database for agent memory and lightweight server-side app state. A local-first note-taking and knowledge-management app for people working in Markdown.

Data model Named JSON tables. Current account limits are 20 tables, 1,000 rows per table, 25 MB per account, and 64 KB per row. Markdown files in a vault, organized with folders, links, backlinks, tags, properties, canvases, and plugin-defined structures.

Query behavior CRUD, simple equality filters, sorting, counting, batch writes, and short-lived viewer links for larger results. Search, backlinks, linked and unlinked mentions, tags, graph navigation, and plugin-powered views over local notes.

Agent interface Hosted Streamable HTTP MCP, local stdio MCP, REST API, and CLI are built into the product surface. Agents can work with Obsidian vaults through file access, plugins, or custom tooling, but the core product is an interactive notes app.

Storage location Hosted on Cloudflare with account-scoped durable storage and API access. Local Markdown files on the user's device, with optional paid services and community workflows for sync, publish, and backup.

Shape of memory Structured rows are easy for agents and apps to validate, update, filter, restore, and share across sessions. Freeform Markdown is easy for people to read, write, edit, link, and reorganize as ideas evolve.

Recovery model Every update and delete archives the previous row. The last 10 versions are kept for 30 days and can be restored. History depends on the user's vault workflow, such as Obsidian Sync version history, Git, filesystem backups, or plugins.

Positioning

A practical way to think about the split.

pumaDB is the memory primitive

It is useful when an agent needs durable state that stays explicit: preferences, task state, project facts, resource references, code snippets, commands, config examples, and small records your own app already understands.

Obsidian is the human workspace

It is useful when the job is writing, researching, journaling, linking ideas, navigating backlinks, reviewing a graph, or building a personal knowledge base in local Markdown files.

Sources

What this comparison is based on.

Product pages change over time. This page compares pumaDB's current documented behavior in this codebase with Obsidian's public project documentation.

Obsidian home Obsidian help pumaDB docs pumaDB install

Start small

Add hosted memory without standing up a database.

Connect pumaDB over hosted MCP, or call the REST API from trusted server-side code.

Install pumaDB Read the API docs