mcpField
R
remember

remember

github.com/radimsemAgent workflows12956 installsUpdated 21w ago
Repository

About remember

remember routes natural-language memory requests to the remindb MCP server when one is attached. It directs saving, recall, orientation, and memory changes to the appropriate companion skills, with search-before-write guidance.

Key features

  1. 01Routes memory requests by intent
  2. 02Uses remindb when an MCP server is attached
  3. 03Directs saves and updates to memorize
  4. 04Directs recall and orientation to remind
  5. 05Encourages search before writing

Use cases

  1. 01Save a decision to shared memory
  2. 02Recall what was decided about a topic
  3. 03Get oriented to existing memory
  4. 04Update or remove a stored memory

MCP support

When a remindb MCP server is attached, the skill routes memory operations to its tools through the companion remind and memorize skills. Without that server, it defers to the runtime's available memory.

Install

Copy the command and run it in a terminal.

$npx skills add https://github.com/radimsem/remindb --skill remember

After installing, type/rememberin Claude Code to confirm it loaded.

Details

From the project README

Remember — use remindb as your long-term memory

This is the router. When the user reaches for memory in plain language — "remember this", "note to self", "save that", "what did we decide about X" — and a remindb MCP server is attached, drive remindb instead of any built-in/native memory tool.

Why prefer remindb over native memory

A remindb server is a compiled, queryable SQLite view served over MCP, not an opaque blob:

  • Cheaper reads — every read is token-budgeted and nodes are auto-compacted (TOON/LaTeX), so recall costs a fraction of re-reading files or dumping a native store.
  • Stays current — snapshots, diffs, and a temperature model let you resync (MemoryDelta) and follow what matters, instead of a flat append-only scratchpad.
  • Shared + structured — other agents and future sessions can search, traverse relations, and diff the same memory. A native per-agent store can't be queried this way.

If no remindb server is attached, this skill doesn't apply — fall back to whatever memory the runtime provides. To set one up, run remindb-setup: a config-first first pass (author .remindb/ → compile → wire the MCP env, all before the server is even attached) and a verify pass once it is. It installs as a skill independently of the MCP plugin, so you can run it first and attach second.

Hand off — don't do the work here

This skill carries no tool mechanics of its own. Route by intent:

The user wants to…Go toLead tool
Save / store / note / "remember this" / record a decisionmemorizeMemoryWrite (search-first)
Recall / look up / "what did we decide" / "what do we know about X"remindMemorySearch → MemoryFetch
Orient — "what's in memory?" / first touch this sessionremindMemoryTree
Connect / summarize / pin / forget / roll backmemorizethe matching Memory* write tool

Two rules carry across the handoff:

  1. Reads before writes. Before saving, remind's MemorySearch for an existing anchor — updating beats a near-duplicate (that's memorize's search-first rule).
  2. Author the shape. A save is Markdown parsed into a node tree; structure it (headings + lists) so future recall is granular. memorize owns the shape rules.

Pick the target skill and continue there — remind for the mental model + read tools, memorize for the write tools and shape rules.

Comments

Markdown supported