Memoshelf vs MemorySync: which memory tool fits a few agents?
Memoshelf (ours) is a shared notebook for a few agents: plain notes, a pinned rules page, a readable run log and page reading, set up without code. If you run a handful of agents and want them to remember, read the same rules and leave a record, Memoshelf is the simpler fit.
Who each is for
Memoshelf fits a small business or solo builder running a few assistants across different apps who wants one notebook they all read and write, without writing code. It does keyword search, not semantic recall, by design.
Side by side
Everything in this table is about Memoshelf (ours). We have not described MemorySync's features or pricing here.
| Memoshelf (ours) | |
|---|---|
| Memory model | Plain notes and a pinned rules page, keyword search |
| Run log | One line per finished run, stale runs shown |
| Page reading | Fetch a public page as markdown, honours robots.txt |
| Integrations | Any assistant that can add a custom connector |
| Enterprise features | Export on every plan; receipts on Team |
| Setup | Sign in, connect, done; no code |
| Free tier | Free: 3 AIs, 500 notes, no card |
| Paid tiers | Solo $15/mo, Team $49/mo |
Set it up for any assistant that can add a custom connector
- Sign in to Memoshelf with your email (free plan, no card).
- Add Memoshelf to your assistant as a custom connector at https://memoshelf.com/mcp and sign in when asked; pick a name for that agent.
- Tell the agent to read what_happened_last at the start of a task and store_note at the end.
- Open the account page to read what every agent wrote.
Try it
Free tier, no card. Related playbook: the entry on an AI that forgets between sessions.
Check it worked: Add the connector, run one task, and read the note it wrote.
New here? Quickstart
Related: