by l33tdawg
SAGE is a governed memory layer for AI agents. Each memory is signed by an agent, admitted through a review step, limited by permissions and kept with an audit trail.
PluginContact for pricing
Step 1: Download the desktop release (macOS, Windows or Linux), or run the Docker image, or build from source.
Step 2: Open SAGE and complete setup in CEREBRUM: choose a memory engine and connect an AI client.
Step 3: Review which agents have access and which domains they can read or write.
Step 4: After connecting an MCP client, call sage_inception and follow the memory mode it returns.
Step 5: Check the node with a request to the local health endpoint, which returns a healthy status when it is running.
The README lists Claude Code, Codex, Cursor and other MCP-capable clients. Developers can also use the Python SDK or the signed REST API.
Yes, by default. Optional federation, connectors and remote model providers can send data to destinations you configure.
No. Its checks govern what is committed and who can use it, but they do not prove every remembered claim is true.
Requirements
Inspeximus MCPInspeximus is tamper-evident long-term memory for AI agents, shipped as one zero-dependency Python file plus an MCP server. A corrected fact stays corrected, and erasures can be proven.
M3 Memory MCPm3 Memory is a local-first memory layer that lets AI coding assistants keep facts, decisions and conversation history across sessions and tools. Assistants reach it through MCP, and it also has a command-line interface.