Your team's AI tools,
finally on the same page.
Shared context for your whole team, across every AI tool they use. Activity, decisions, and tasks across all five layers keep every teammate's AI current. Claude, Cursor, ChatGPT, and any MCP client read it and write back automatically.
Your context shouldn't live in one vendor's app.
The questions your tools can now answer
“What did we decide about the pricing model, and why?”
Response
From the ledger: you moved to simple per-user pricing in the Q1 review and parked usage caps and per-action billing, because most accounts are small teams. The rationale and who made the call are attached.
The context graph
One graph. All context.
Your whole company's context, woven into one living graph around BaseThread. Hover a hub to trace its world.
BaseThread connects five levels of context around your workspace, each holding what your AI tools read:
- Company: Mission, Vision, Brand voice, Product strategy, Target personas, Competitors, H1 roadmap, Candidates called, Customer base, Tech stack.
- Products: Product decisions, Active bets, Pricing, Positioning, OKRs, Architecture decisions, Release notes, Feature specs, Shipped this week, Integrations.
- Teams: Members, Tasks assigned out, Process decisions, Team goals, Tooling decisions, Pipeline decisions, Domains owned, Posts lined up, Leads followed up, Comms channels.
- Projects: Brief, Success metrics, Deliverables, Owners, Decisions, Open questions, Activity, decisions & tasks, Open tasks, Launch checklist, Risks.
- You: Your role, Preferences, How you work with AI, Current focus, What you own, Assigned to me, Working hours, Recent decisions, Goals, Communication style.
Shared with everyone
Every person and every AI tool in the company reads from the same graph, so no one re-explains the same context twice.
Activity, decisions & tasks
Each teammate's AI logs what shipped, what was decided, and what's next, harmonized in the cloud and pushed to every member's tools.
Owners stay in control
Multi-tier permissions keep each layer's owners holding its source of truth, with the right access for everyone else.
The glue
Every ship, every decision, every task,
on the same page
Activity, decisions, tasks, and the content across all five layers keep every teammate's AI current, and get sharper every day.
No rules to write. No projects to set up.
Your AI does the remembering.
Each member's local AI summarizes their work and writes activity, decisions, and tasks to the BaseThread cloud. The cloud harmonizes them and pushes the latest back to every member's AI, so the whole team stays in sync.
Each member's AI writes back
Back as three streams, to every member
Activity, Decisions & Tasks
Each member's AI is the scribe
As work happens, your local AI summarizes activity, extracts decisions, and tracks what's next, then writes all three back. Every other tool reads them. No status meetings, no manual logs, no stale handoffs.
ledger.activity({ text: ["Shipped the new onboarding flow"], affects: ["Product", "Growth"], by: "sarah's Claude Code" })
ledger.decide({ text: ["Default new workspaces to Team mode"], rationale: ["Most signups are teams, not solo"], affects: ["Product", "Growth", "Support"], by: "maya@acme.com (via Claude)" })
ledger.task({ text: ["Draft the launch post"], assignee: "priya@acme.com", due: "Fri", by: "maya's Claude" })
Two surfaces
A platform to manage your source of truth. A native app to keep every machine current.
The platform
Where your team collaborates on, maintains, and manages the source of truth: your layers, projects, members, and decisions, all in one place.
Native Mac app
The utility that keeps every member's machine current, so the AI tools running locally on it always have the full context they need.
Your AI already remembers, just for you, in one app.
BaseThread is the shared context for your whole team, across every AI tool they use.
How it works
Context in once. Read everywhere. Written back automatically.
Define context once
Your team writes its company, products, projects, and decisions into one structured place.
Sync to every machine
The BaseThread agent mirrors that context to every member's machine for fast, offline reads.
Every tool reads it
Claude Code, Cursor, and any MCP tool pull the right context at the start of every session.
Agents write back
When an agent finishes, it logs what shipped, what was decided, and what's next, so others stay caught up.
Why teams run on it
The things you only get when context is shared, structured, and live
Onboard people and agents in seconds
A new teammate or a fresh agent session starts already caught up on your decisions, conventions, and current work. No ramp-up.
Personal and team, one system
Run it for your whole company, or just for yourself and your family. Same architecture, the terminology adapts.
The right context, not your whole company
Tools pull only what's relevant to the task instead of stuffing everything into the prompt, so answers stay sharp and you don't burn tokens.
Activity, decisions, and tasks, all queryable later
Activity ('what did Sarah ship this week'), decisions ('what did we settle on pricing, and why'), and tasks ('what's on my plate, and what did I delegate') all live in one place, so any agent can answer, and assign, months later.
Security & privacy
Better, smarter AI context, and it stays yours.
Local-first
Your context syncs to your own machine for fast, offline reads.
Encrypted at rest
AES-256-GCM on sensitive context, with the key outside the database.
Your tools, your accounts
Connected AI tools run on your own accounts; BaseThread just serves them context over MCP.
Never shared or trained
Never leaves your workspace, never used to train models.
Get your team's AI tools on the same page
BaseThread is the shared context-graph that Claude Code, Cursor, and every AI tool your team uses can read, so no one re-explains the same context twice.
Request access