MCP SERVER · v0.1.0 · BETA

Give your agent
a memory forest

AllMyMind · 记忆森林

A leaf is something you said. A branch is how it connects. A tree is a topic that has grown. What should fall, falls — because it forgets clearly.

FormsTree · Shrub · Grass
EntryCLI · HTTP · MCP
Versionv0.1.0
01 / Why

Most agents meet you for the first time, every time

Close the context window and everything is gone. Paste the whole history in and attention dilutes. A window is not a memory.

01

Amnesia

Yesterday's decisions, your stack preferences, the unwritten rules of the project — gone when the session ends.

So it recommends the approach you already rejected, and asks what you answered three times.

02

Bloat

Cramming history into the prompt inflates tokens and drowns the one line that mattered.

Costs go up, quality goes down.

03

Opacity

Even when it remembers, it cannot say why this came up. Nothing to debug, nothing to reuse.

Unexplainable memory cannot be trusted — or improved.

02 / Three laws

Three laws of the forest

AllMyMind is not a wrapper around a vector store. It answers “what should memory look like” before “where do we put it”.

Structure before retrieval

Memory is not a pile of loose vectors. Trunks are themes, branches are threads, leaves are what you actually said.

Grow the tree first, then look for leaves.

Forgetting is a feature

Grass untouched for two days withers; shrubs for four. Leaves fall into the fallen layer — still findable, but out of the way.

You have to know what to forget before you know what is worth keeping.

Explainable recall

Every “I remember” can name the branch it hangs on, when it was stored, and why it surfaced now.

Memory matters when it can be questioned.

03 / Forms

What it remembers for you

Leaves, branches, trees, shrubs, grass — each has explicit rules. Code decides what shape it becomes.

Leaves

One sentence, one decision, one excerpt. Lands in the pending layer, waiting to be placed.

Branches & trees

Three solid branches and three leaves each make a tree. If it cannot hold, it falls back to leaves.

Semantic search

It searches meaning, not keywords: “remote access fails after locking” finds the RustDesk headless note.

Digging

Drop a whole folder in. Inspect first: structural reading costs zero model calls, garbage is refused.

Hollows

Attached to trunks only, excluded from search. Not even “there is a memory here” is revealed.

Continued conversation

Elsewhere, it offers a card first: how many, how long, last stored. You approve before any text is sent.

04 / Connect

Three commands, one leaf

Data lives in local SQLite. Models are used only for digging and grouping — it runs without a cloud.

MCP needs PYTHONPATH in the env block — see the docs.

Read the docs
host config · mcp.json { "mcpServers": { "allmymind": { "command": "npx", "args": ["-y", "@allmymind/mcp"], "env": { "ALLMYMIND_HOME": "~/.allmymind" } } } }
05 / Updates

What is growing

All releases
2026-11-03 · v0.1.0

First usable build

Four tables plus local Chroma; shape rules (tree / shrub / grass) enforced in code.

2026-10 · v0.0.9

Handoff

Manual pipeline: select a tree, “send to conversation”, packaged and copied to clipboard.

2026-10 · v0.0.6

Digging

inspect → run → apply, with whole-batch rollback on failure.

Plant the first tree

AllMyMind is in beta. If you are building agents too, come try the first patch of forest.