Open source · Works with Claude, Cursor, Kiro & more

Your AI forgets.Knowns remembers.

Turn tasks, specs, docs, templates … into live context for your AI assistant — with structured memory across sessions.Fresh project state anytime.

Incremental
Structured context
Session memory
knowns · AI workflow
LIVE
Context loaded
task-42
@doc/auth-spec
@template/api-auth
JWT pattern
4
nodes
Δ 1
delta
Y
Work on task 42
Tell your AI what to work on…
188
GitHub Stars
2.3k+
Downloads/mo
40
Forks
MIT
License

Everything your AI needs to understand your project

Structured context, not scattered files. Four pillars that turn your AI assistant from a code generator into a project-aware collaborator.

Project Memory
Your AI doesn’t start from zero. Decisions, conventions, and patterns persist across sessions.
  • Project memory for repo-level patterns and decisions
  • Global memory for cross-project conventions
Structured Execution Context
Tasks with acceptance criteria, linked specs, and templates. Your AI knows what “done” looks like before writing code.
  • Tasks with ACs that sync to specs
  • Spec-driven development workflow
  • Code generation templates linked to docs
AI Access Layer
MCP server, semantic search, and code intelligence let your AI find the right context without you pointing at files.
  • MCP integration for Claude, Cursor, Kiro & more
  • Semantic search with native ONNX embedding
  • Code graph with symbol search and dependency tracking
Local-First Workflow
Native Go binary. Everything on your machine. Web UI with kanban, docs editor, and AI chat built on OpenCode.
  • Fast native binary with built-in ONNX embedding
  • Browser-based workspace with kanban, docs, and AI chat
  • Import and sync docs/templates from Git repos or npm
  • Workspace switching across multiple projects

And more

Import & Sync
Import docs and templates from Git repos, npm packages, or local folders. Keep them synced.
Semantic Search
AI-powered search using native ONNX vector embeddings. Find tasks and docs by meaning, not just keywords.
Code Intelligence
Search symbols, inspect code relationships, and explore a code-aware graph without leaving your local workflow.
AI Chat with OpenCode
Multi-session AI chat with real-time streaming, model switching, and optional task linking.
Workspace Switching
Switch projects in the browser while keeping kanban, docs, memory, graph, and AI chat close at hand.
Validate & Quality Checks
Check tasks, docs, and templates for broken refs, missing ACs, and SDD coverage gaps.

See the full feature set on GitHub →

Incremental context flow

Context flows through your project.
Only the delta updates.

When a task, spec, or template changes, Knowns traces the impact downstream. Your AI always works with fresh, connected context.

Δ AC
Task
task-42
Spec
@doc/auth-spec
Template
@template/api-auth
Code
src/auth/middleware.ts
Memory
JWT pattern saved
AC updated → re-validate downstream
5
Context nodes
linked
Δ 1
Delta updates
this change
4/5
Cache hits
reused
Web UI · Built-in

See your project state.
In real time.

Kanban board, docs editor, AI chat, and semantic search — all in your browser. Run knowns browser

localhost:6420/#/kanban
LIVE

Task Board

6 tasks · 2 specs linked
To Do2
Add rate limiting
task-45
medium
Write API docs
task-46
low
In Progress1
User authentication
task-42
SPEChigh
In Review1
Database schema
task-40
SPECmedium
Done2
Project setup
task-38
low
CI/CD pipeline
task-39
medium
Video Demos

Workflow Demos

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

Task Workflow
~2 min
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 →

You might be thinking
“I already have something for this”

“I use CLAUDE.md / AGENTS.md”

Those are static instruction files. Knowns gives your AI structured tasks with acceptance criteria, linked specs, searchable memory, and a code graph. Instructions tell AI how to behave. Knowns tells AI what to build and what “done” means.

“I use a task manager / Task Master”

Task managers track work for humans. Knowns structures tasks so AI can read them, link them to specs and templates, verify acceptance criteria, and extract knowledge when done.

“I have project docs”

Docs are great. Knowns makes them searchable by meaning (semantic search), linkable to tasks and templates, and accessible to your AI via MCP. Your docs become part of the execution context, not just reference material.

“I already use MCP tools”

MCP is the transport layer. Knowns is the structured context that travels through it. Tasks, docs, specs, memory, templates, code intelligence — all exposed through one MCP server.

“I use SpecKit / spec tools”

Spec tools help you write specs. Knowns connects specs to everything else: tasks link to specs, ACs sync automatically, templates generate code from specs, and verification checks coverage. Specs aren’t standalone documents — they’re nodes in your project graph.

“I use Claude-Mem / memory tools”

Memory-only tools store facts. Knowns structures them: project-level vs global, tagged and searchable, linked to tasks and docs. Memory is one layer — Knowns connects it to specs, tasks, templates, and code intelligence so your AI acts on memory, not just recalls it.

Get started in under a minute

Install Knowns, initialize in your project, and connect your AI assistant.

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
# 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

Open Source

Built by the Community

Knowns is open source and made better by contributors like you. Join us in building the future of knowledge management.

Report Issues

Found a bug? Open an issue to help us improve.

Submit PRs

Have a fix or feature? We welcome pull requests.

Improve Docs

Help make our documentation clearer and more helpful.

Our Contributors

Meet the Team

The people building Knowns to help developers manage knowledge better.

HN

Howz Nguyen

Founder & Lead Developer

HV

Harry Vu

Developer

Interested in contributing? Join us on GitHub

Join the Community

Connect with developers building with Knowns. Share tips, get help, and stay updated on new features.

Get Help

Ask questions and get answers from the community

Early Access

Be the first to try new features and updates

Connect

Meet other developers using Knowns

Free to join. No spam. Just developers helping developers.

Knowns

Stop managing your AI’s memory.
Start building.

Give your AI assistant structured access to your project. Local-first. Open source. Free.

npm install -g knowns

Open source · Self-hosted · MIT License