
Context Engine
One memory for your AI agents and human team members.
Sign in to rate this tool.
Overview

Context Engine is a shared memory hub for AI tools and teams. It lets Claude, ChatGPT, Cursor, and Claude Code store and retrieve information across sessions. The service is free to start and requires no credit card.
When you sign in with your email, the first step is to connect a tool. For ChatGPT and Claude the process is a few clicks; for Claude Code it is a single command that points to an engine's URL. Each connection is limited to the engines you tick, and the system does not expose any keys to users. Connected tools then use the MCP protocol to push content into the shared database. The example on the demo page shows the invoicing app writing a decision and uploading a .zip of a PDF; the data appears in the web interface with the tool name and a timestamp.
On startup of a tool, it receives a short briefing that lists recent activity across all engines it can see. In the demo, the briefing includes a statement about a decision to limit retries, a note from Cursor about cost‑centre grouping, and a new contact for Contoso. If the tool needs more context, it can ask the memory for a matching record, and the system searches using both exact words and synonym matching. The memory is organized by “engines,” each defined by a one‑sentence purpose that tells a tool where to save or pull data. A user can create one engine for themselves, one per client or project, or one for the whole team, and permissions can be set to read‑only or read‑write.
The web app shows every memory in plain text, tagged with its type (decision, note, person, file) and the source engine. Users can drop in individual files, whole folders, or zip archives. All exported data can be downloaded as Markdown or cleared entirely, giving full ownership of the shared knowledge.
This tool is useful for teams that use multiple AI assistants and want to avoid re‑exposing the same context to each. It is less suitable for workflows that require heavy custom integration or for users who need to keep data strictly offline, since the service is hosted and relies on MCP.
Details
- Pricing
- Free
- Last checked
- 17h ago
- Indexing
- noindex, follow
Suggest an edit
Sign in to suggest an edit.
Pros and cons
Pros
- Unified context eliminates repetitive explanations across tools
- No key management; connections are short and secure
- Team‑wide visibility of AI outputs reduces silos
- Transparent record of who saved what and when
- Supports file uploads, including folders and zip archives
Cons
- Free plan does not disclose storage limits or rate limits
- No publicly available pricing information
- Requires manual connection for each tool
- Lack of details on offline or on‑premise deployment options

