Automatic memory capture
Supported workflows capture useful project context in the background. Important decisions can still be saved explicitly when needed.
Gitmas Memory · smart shared recall
Every agent. Every vendor. One memory.
Gitmas Memory is one centralized brain for agents across different vendors and models – across sessions, machines, restarts, and projects. Encrypted with your key, private by design, recallable anywhere. What one agent learns, every agent remembers.
CLI, MCP, and API access
Use Gitmas through supported local tools, MCP clients, and API integrations to recall context whenever the work needs it.
Encrypted session saves
Autosave archives full sessions encrypted, every minute – closing a terminal is never the moment that decides whether your work was saved. Your memory is not sold or used to train models.
Zero-token memory
Storing, syncing and archiving happen outside the model – no tokens spent. Recall is on-demand: one short memory hit typically replaces thousands of tokens of re-reading and re-explaining. Shared memory doesn’t inflate your AI bill – it shrinks it.
Rules your agents can’t forget
Store a rule once – every agent, from every vendor, on every device gets it injected at session start. Hard rules stop depending on recall.
Readable session archive
Browse sessions in the dashboard, read full transcripts with highlighted code and in-text search, and jump from any checkpoint to the exact moment it describes.