USECREA
Project memory for AI agents & developers.
USECREA gives your AI agent a persistent, structured brain. Every decision, task, insight and dataset lives in one place — readable and writable by agents and humans alike, across sessions, tools and time.
Development Timeline
Log every step as a versioned entry. Each entry carries a stage, visibility and free-form description — a living changelog your agent can read and extend.
Knowledge Base
Store decisions, insights, test results and hypotheses. Link them together, pin the critical ones, mark relevance and let agents supersede outdated knowledge.
Task Tracking
Create tasks with priorities, blocking dependencies and subtasks. Agents update status automatically — open, in progress, blocked, done or cancelled.
Project Elements
Register the components of your project — libraries, services, APIs, modules — with type, description, URL and custom properties.
Raw Data Storage
Persist chat histories, code snapshots and structured datasets across sessions. Agents save and retrieve context without losing state between conversations.
File Storage
Agents upload and retrieve files — code, images, reports — directly via API. Files are attached to timeline entries and accessible across sessions.
API Key Permissions
Issue multiple keys with fine-grained control: read, write and delete permissions independently, scoped to specific projects. Full access by default.
Full Export
Download a complete ZIP of any project at any time — metadata, timeline, knowledge, tasks, datasets, snapshots, chat sessions and all uploaded files.
MCP tools — for any agent framework
Copy a ready-made connect command or config for your agent. Agents get tools to start a working session, read and write all layers, search across everything and close the session with a structured summary — so the next agent can continue seamlessly.
Works with
- —Claude Code
- —Claude Desktop
- —Cursor
- —Grok CLI
- —Grok
- —Any MCP client
Connect your agent in minutes
Four steps — from signup to your first session.
Create an account
Sign up for free. No credit card required. Your projects, keys and data are yours.
Generate an API key
Go to Settings → API Keys. Create a key and set permissions — read, write, delete — per project if needed.
Connect your agent
Run the MCP connect command in your Agent or CMD, paste the config — done.
Start your first session
Tell your agent to start working on a project. It calls start_working, loads the full context and is ready to go.
Use natural language prompts
Use the USECREA keyword in your agent – USECREA handles the rest via MCP.
Start a new project
“Create a new USECREA project called "My App". Set a short description, add the main tech stack as project elements, and create 3 starter tasks.”
Continue where you left off
“Start working on my USECREA project "My App". Give me a full briefing — current status, open tasks, recent decisions — and tell me what to focus on next.”
Document a decision
“We just decided to use PostgreSQL instead of MongoDB as the main database. Save this as a decision in the USECREA knowledge base with the reasoning and any relevant trade-offs.”
Close a session
“We're done for today. Create a timeline entry for what we accomplished, mark the finished tasks as done, save any open decisions in the knowledge base, and end the session with a summary.”
Search & summarize
“Search my USECREA project "My App" for everything related to authentication — decisions, tasks, entries and knowledge — and give me a structured summary.”
Save a snapshot
“Take a snapshot of the current project architecture and save it in USECREA so we can compare it to the state in a few weeks.”
Ready to give your agent a memory?
Create your first project