Open source project memory for teams and AI

Your project remembers now.

Every task, decision, and pattern stays connected.

Knowns keeps tasks, docs, and decisions linked, searchable, and ready for every coding agent.

Community support, setup help, and release notes live in Discord.
Knowns Web UI dashboard showing project status, task metrics, charts, and recent activity.

Tasks

Linked to the same project state.

Docs

Linked to the same project state.

Memory

Linked to the same project state.

Real Knowns web interface with project tasks, docs, and memory in one workspace.
214
GitHub Stars
1.3k+
Downloads/mo
MIT
License

One memory loop.

Knowns keeps the working state readable for people and coding agents.

Capture context

Bring tasks, docs, decisions, and context into one place.

Link work

Connect related work, patterns, and dependencies.

Verify state

Validate with your team, agents, and automated checks.

Reuse memory

Institutional memory that stays accurate and useful.

Actual Web UI

A real workspace for tasks, docs, memory, and impact.

Knowns is not just a CLI surface. The browser UI gives teams and agents the same project graph with board views, task detail, documentation, and relationship maps.

Included surfaces

DashboardKanbanTasksGraphDocs

Project dashboard

A command center for progress, status, task flow, and project activity.

Kanban board

Move work through status columns while keeping every task tied to project context.

Task detail

Open a task with acceptance criteria, references, notes, and progress in one focused view.

Context graph

See how tasks, docs, memories, and decisions connect before changing the system.

Docs workspace

Browse project documentation next to the same graph and task relationships.

Context disappears. Work repeats.

Knowns turns scattered project traces into a working memory layer your team and agents can read.

Without project memory

  • Tasks live in one tool, docs in another, decisions in chat
  • Every session starts with re-explaining the project
  • Handoffs lose context — new people ask the same questions
  • AI assistants guess because they can't see the full picture
  • Good solutions get forgotten and reinvented

With Knowns

  • Tasks link to docs, specs, and decisions in one place
  • Sessions start with full project state — no pasting needed
  • Handoffs include history, decisions, and linked references
  • AI reads structured context and knows what 'done' looks like
  • Knowledge is extracted and reused through templates

From idea to done, with context intact.

A simple cycle that keeps knowledge connected.

1

Capture

Write down what needs to happen and what 'done' means.

2

Link

Connect tasks to docs, specs, templates, and past decisions.

3

Work

Your AI reads the full context and follows the plan.

4

Verify

Check acceptance criteria before marking complete.

5

Remember

Extract patterns and decisions for next time.

Video Demos

Workflow Demos

Real-world examples of using Claude Code with Knowns skills.

Claude CodeFull Workflow

Task Workflow

Use /kn-research, /kn-plan, /kn-implement, and /kn-extract to implement dark mode in a Flutter project.

All skills use the /kn-* namespace. View full workflow guide →

Everything you need to keep context

Six capabilities that work together. One connected system.

Tasks

Plan work with acceptance criteria, linked specs, and progress tracking.

Docs

Keep project knowledge searchable, versioned, and close to work.

Templates

Reuse proven workflows instead of starting from scratch.

References

Explicit links between tasks, docs, and decisions. Nothing orphaned.

Search

Find anything by meaning, not just keywords. Semantic search built in.

Agent Context

Give AI assistants structured project memory via MCP integration.

Built for real project handoffs

Knowns helps people and AI agents start from the same context.

Solo Developer

You build projects across multiple sessions. Knowns remembers your architecture decisions, tracks what's done, and gives your AI the full picture every time you start working.

AI-Assisted Team

Multiple people and AI agents working on the same codebase. Knowns ensures everyone — human or AI — works from the same structured context. No conflicting assumptions.

Product & Engineering Handoff

Specs, tasks, and acceptance criteria live in one connected system. When engineering picks up a task, the full context is already there — linked docs, related decisions, and clear definition of done.

How Knowns compares

Most tools cover one slice. Knowns connects them all.

 KnownsBMAD MethodSpec KitBacklog.mdmem0Claude MemoryCursor / Windsurf rulesRepomix
Structured tasks with ACsStory-likeSpec tasks
Linked specs & docsPartialSpecs only
Semantic searchPartial
Project memory across sessionsProcess onlyStatic rules
Templates & code generationRule templates
AI agent context (MCP)Via integrationsThird-partyHooks
Validation & coverage checksProcess QAPartial
Web UI (kanban, docs, chat)API only

Get started in under a minute

Install, initialize, and your AI assistant is connected.

Prerequisites

Multiple install paths are available including Homebrew, shell script, PowerShell, npm, and `npx`. npm-based installs still need Node.js 20+. Download Node.js

Install with Homebrew
$ brew install knowns-dev/tap/knowns
Install with Shell
$ curl -fsSL https://knowns.sh/script/install | sh
Install with PowerShell
$ irm https://knowns.sh/script/install.ps1 | iex
Install with npm
$ npm install -g knowns
Install with npx
$ npx knowns
Quick Start
# Install with shell script
$ curl -fsSL https://knowns.sh/script/install | sh
# Or install with PowerShell
$ irm https://knowns.sh/script/install.ps1 | iex
# Initialize Knowns in your project
$ knowns init
# Setup AI platform integrations (global)
$ knowns setup claude --global # or: knowns setup all --global
# Or try without installing globally (npx)
$ npx knowns init
# Verify installation
$ knowns --version
# Upgrade later without reinstalling
$ knowns update
# Start the local UI + AI Chat (optional)
$ knowns browser --open

Keeping Skills & Agents Updated

After a new release, update your installation then run:

knowns sync

This keeps your skills and agents always up to date.

Platform-Specific Instructions

Give your project a memory.

Start with one project. See the difference in your next AI session.

Open source · Local-first · MIT License · No vendor lock-in