Practical overview
When it fits
Use BrainX when OpenClaw agents need a persistent, searchable memory store or governed memory sharing across agents, and you can operate its database and embedding prerequisites.
Who it is for
- OpenClaw developers who want agent memories to persist across sessions.
- Teams operating multiple OpenClaw agents that need to share verified memories.
- Operators who want CLI-based storage, search, diagnostics and memory maintenance.
Documented uses
- Store decisions, corrections and operational learnings for later sessions.
- Retrieve memories by meaning rather than exact keywords.
- Inject relevant stored context into an agent prompt.
- Share verified, governed memories across agents.
- Optionally connect memory to OpenClaw runtime surfaces such as wiki digests, working memory and JIT recall.
Requirements & limits
The repository is public and displays an MIT license badge. The README lists local software, database and OpenAI API key requirements.
- The CLI is the durable memory engine; the OpenClaw plugin is an optional runtime adapter.
- The documented stack requires PostgreSQL 14 or later with pgvector, Node.js 20 or later, OpenClaw and an OpenAI API key.
- The README recommends enabling runtime write-path features gradually, after diagnostics and review of retention and privacy policy.
- The README says memory is advisory and should yield to current code, logs, tests, database state and direct user corrections.
Check current requirements in the official documentation. Source-based guidance does not mean OpenViST has run or independently tested this project.
Set up BrainX’s local memory engine
Prepare the documented BrainX CLI and PostgreSQL memory store.
Before you start
- OpenClaw
- Node.js 20 or later
- PostgreSQL 14 or later with pgvector
- An OpenAI API key
Steps
Clone the BrainX repository, install npm dependencies and make the brainx CLI executable.
Create a PostgreSQL database and enable the vector extension.
Apply the SQL schema and migration files.
Copy the example environment file and configure DATABASE_URL, OPENAI_API_KEY, the embedding model and embedding dimensions.
Run the documented doctor and health commands, and the npm test command.
Expected outcome
The repository documents a configured memory store and CLI checks for diagnostics and tests.
Guide limitations
- These are repository setup instructions, not a guarantee that setup or tests will succeed in a particular environment.
- This guide does not configure the optional OpenClaw plugin.
Add and retrieve a memory with the CLI
Store a decision and retrieve related context using the documented BrainX CLI.
Before you start
- A configured BrainX CLI and database
- The OpenAI API key and embedding settings required by BrainX
Steps
Use the documented brainx add command to save a memory, with its type, importance and context options.
Use brainx search with a query and result limit to find semantically related memories.
Use brainx inject with a question to retrieve context for an agent prompt.
Expected outcome
The CLI commands add a memory, search for related information and return context for injection.
Guide limitations
- The guide covers CLI operations only; it does not configure runtime plugin behavior.
- Search results and successful command execution depend on the local installation and stored data.
Sources & review
Source record details
- Source availability
- Active
- Sources checked
- 18 October 2026
- Original website title
- Mdx2025/BrainX-The-First-Brain-for-OpenClaw · Official homepage title · retained in its source language