KnowIt | Open-source memory for AI coding agents

Scattered notes vs. one place agents ask.

Rules and markdown pile up. Each tool's memory stays in that tool. Knowit gives agents structured, queryable answers without a growing doc pile.

Before

Files and rules multiply; each app remembers its own session.

After

One queryable layer. Every agent reads it the same way.

Stop re-explaining your codebase.

Without Knowit

Your agent asks how Stripe webhooks should be handled. You explain. Next session, same question.
Another developer's agent introduces a naming pattern you deprecated six months ago.
Three agents working on the same codebase hold three inconsistent models of the architecture.
The repo keeps growing new memory markdown files because no one knows where durable agent context should live.

With Knowit

Agent calls resolve_context before planning. It finds the idempotency rule. It starts informed.
Every convention is immediately searchable by every agent on the project.
At session end, the agent persists everything it discovered.
Code stays in the repo. Durable memory moves into Knowit instead of becoming one more markdown file.

“The problem is not your agents. It is that their memory resets every session.”

Set up in three steps.

Run the install wizard.

Run npx knowit install inside your project. The wizard picks your client, scope, and storage, then wires everything up. It can install knowit globally for faster future starts, or skip it and use npx.

npx knowit install

Your agent picks up Knowit automatically.

Instruction files and MCP registration are updated for you. Agents call resolve_context before planning and capture_session_learnings after, with no manual wiring needed.

# agent workflow, automatic
resolve_context("add retry logic")
# ... agent works ...
capture_session_learnings()

What you get after setup.

Your project now has persistent memory wired into the agent workflow: client registration, project instructions, optional project MCP config, and migrated knowledge from architecture or ADR-style docs.

# after npx knowit install
✓ agent connected to Knowit MCP
✓ project instructions updated automatically
✓ project MCP config added for compatible clients
✓ architecture / PRD / ADR docs moved into Knowit memory

Shared context, not solo sidecars.

Structured knowledge model

Types (rule, architecture, pattern, decision, convention, note), scopes (global, team, repo, domain), and confidence scores. Durable memory stops living as loose repo markdown.

Shared memory

One database, every agent on the project. What one session captures, the next session and other agents can read immediately.

MCP-native

Works with Claude Code, Codex, Cursor, and any MCP-compatible agent. A first-class MCP server, not a wrapper.

Less repo markdown clutter

Keep source trees focused on code. Store durable memory in Knowit instead of adding more AGENTS, architecture, decision, and pattern docs to every repo.

Source orchestration

Connect Notion or another MCP source. Knowit's routing layer tells your agent when to use local memory directly and when to call a downstream provider next.

Local-first

SQLite. Knowledge lives on your machines and your infra. No account, no cloud dependency, no data leaving your control.

Built-in memory browser

Run npx knowit preview to open a local read-only UI. Browse, search, and filter everything in your knowledge store — no cloud account needed.

Public launch, open source first.

The local-first release is live now. Cloud plans are paused while we gather feedback, stars, and early contributors.

Open Source

$0

Available now

✓Local SQLite
✓All MCP tools
✓Unlimited entries
✓MIT licensed
✓CLI access
✓Memory preview UI

Read the repo

Pro

Soon

Planned after launch feedback

✓Cloud database
✓Semantic search
✓API tokens
✓Cross-device access
✓Early adopter feedback shapes the scope

Request it

Team