Brainfork gives OpenClaw agents persistent memory — decisions, documents, and context that survive restarts and carry forward session to session.
openclaw plugins install @brainfork/brainfork-openclawYour agent has no memory of yesterday's decisions, yesterday's work, or anything you told it last time. You repeat yourself. Every. Single. Time.
Brainfork stores documents, decisions, and retrieval context in a hosted memory layer that stays portable and inspectable. OpenClaw gets a first-class memory plugin. You keep the source of truth.
Every session begins with you re-explaining the project, the decisions, the conventions. Wasted time. Wasted patience. Every. Single. Time.
You chose Postgres over SQLite for a reason. Your agent doesn't remember. It suggests SQLite again tomorrow. And the day after.
MEMORY.md grows stale. Daily notes pile up. No agent reads all of it. You're the only one who holds the full picture.
Agents should remember your work without owning your memory. Brainfork stores it externally, retrieves what's relevant before each session, and keeps your knowledge sovereign — under your control.
Recall what matters
Inject relevant context before the agent starts — without dumping your entire archive into the prompt.
Sync real working memory
Push MEMORY.md and memory/**/*.md into Brainfork with deterministic hashes, updates, and cleanup.
Keep memory sovereign
Your knowledge stays in a Brainfork server you control, preserved beyond any single model vendor.
Create your Brainfork server
Sign up, create a server and get your OAuth key — takes under two minutes.
Install the OpenClaw plugin
One command: openclaw plugins install @brainfork/brainfork-openclaw. Point it at your server.
Memory syncs automatically
After every session, your agent's memory files are pushed to Brainfork. Before every session, relevant context is recalled and injected.
Decisions are logged and retrievable
Every significant decision your agent makes is captured, tagged, and retrievable — building an auditable history of your project.
Install the plugin, point it at your Brainfork server, and let it push documents, log decisions, and recall the right context before each run.
1
Create a Brainfork server and token.
2
Install the plugin and set the memory slot.
3
Let OpenClaw sync MEMORY.md, recall context, and keep decisions durable.
What is Brainfork?
Brainfork is a hosted memory layer for OpenClaw agents. It stores your agent's working memory, decisions, and documents — and recalls the right context before each session starts.
Which AI tools work with Brainfork?
Brainfork is built specifically for OpenClaw. It uses the OpenClaw plugin system to hook into agent sessions automatically. Standard MCP tools (Claude, Cursor, ChatGPT) can also connect to the knowledge base.
Is my data secure?
Your data stays in a Brainfork server you control. We don't use your data to train models. You can export or delete your memory at any time from your dashboard.
Do I need a credit card to try it?
Yes — we require a card to start the 14-day trial, but you won't be charged until the trial ends. Cancel any time before then and pay nothing.
What happens if I go over my limits?
Overage is €1 per 10k extra requests and €1 per 250 extra indexed documents. You'll always see usage in your dashboard before any overage charge hits.
Can I use Brainfork with multiple agents?
Yes. Each agent workspace syncs its own memory independently. You can have multiple agents (Gertrude, Osborn, Neville…) all writing to and reading from the same Brainfork server.
Pricing
Includes 25k requests and 1k indexed documents each month. Overage is €1 per 10k extra requests and €1 per 250 extra indexed documents.