Human-curated, project-scoped context for the current task — not another generic vector store. Capture what matters, organize by project, and connect agents over MCP.
No credit card required · Works with OpenAI, Anthropic, Gemini
Plays nicely with the models you already use
The problem
Every prompt is a fresh start. You pad the context with history, hit token limits, watch quality drop, and burn money on tokens the model doesn't need to see.
Stuffing chat history into every request wastes tokens and slows responses.
Users repeat themselves across sessions. The model never learns who they are.
Notes, docs, and decisions live in five different tools — none of which the LLM can see.
Features
Tasks, curated context, and MCP tooling that keep agents aligned with how your team already works.
Notes, docs, links, and decisions you choose — not auto-ingested chat sludge.
Agents load the active task and its scoped context, session after session.
Organize by project or bucket. No bleed between tenants or workstreams.
Drop-in MCP server for Cursor, Copilot, Claude, ChatGPT, Codex, and HTTP clients.
Row-level security, encryption at rest, and full audit trail.
Fast reads so getTaskContext and search stay snappy in the coding loop.
How it works
A project-centric workflow built around the current task — curated by humans, consumed by agents.
Capture
Attach notes, docs, links, and decisions to a project or task — human-curated context, not a dump of chat embeddings.
+ Add context — you choose what agents may see
Pricing
For teams shipping AI features.
For regulated and high-scale workloads.
Spin up your first memory store in under a minute. No credit card, no setup calls.