Skip to main content

Documentation Index

Fetch the complete documentation index at: https://docs.maximem.ai/llms.txt

Use this file to discover all available pages before exploring further.

​
What is the MACA?

The Memory Architecture Configuration (MACA) is the per-instance configuration that controls how memories are extracted, stored, and retrieved for your agent. Among other things, it tunes which memory categories are captured, how memories are partitioned across users and customers, which retrieval modes are enabled, and how aggressively long conversations are compacted. The MACA is generated automatically by Synap. You do not write it by hand.

​
How the MACA is created

1

You describe the agent

When you create an Instance you upload a Use-Case Markdown file — a short document describing what your agent does, who its users are, what tasks it handles, and any sensitivity or compliance constraints.
2

Synap generates the MACA

Synap analyzes the use-case file and produces an optimized MACA tailored to that agent — picking sensible defaults for memory category extraction, scoping, retrieval, retention, and ranking based on the agent’s domain and audience.
3

The Instance runs on it

Every memory ingested into the Instance and every context fetch is governed by the active MACA. You can view the current configuration in the Synap Dashboard under your Instance’s settings.
4

It evolves with your use case

Re-upload the use-case file whenever your agent’s purpose materially changes (new task categories, new compliance requirements, new audience). Synap re-evaluates and updates the MACA. The previous version is preserved so you can roll back if needed.
The MACA is an internal serialization — its field names and structure are subject to change as Synap improves its memory pipeline. The contract you depend on is the use-case file format and the SDK / API surface, not the raw MACA.

​
Why we generate it for you

Memory configuration is a deep optimization problem: the right embedding model, scope partitioning, chunk size, ranking weights, and retention policy depend on what kind of agent you’re building and what kind of conversations it has. Hand-tuning every knob is the kind of work that delays a real launch by weeks. Synap’s job is to handle that. Your job is to describe the agent.

​
Next steps

Writing a Use-Case Markdown File

The template Synap uses to generate your MACA. The more detail you provide, the better the configuration.

Memory Scopes

How memories are partitioned across users, customers, and clients.

Memory Types

Which categories of structured knowledge Synap extracts.

Managing Instances

View the current configuration and re-upload your use-case file in the Dashboard.