SaveSlot Docs

Account

FAQ

Does SaveSlot answer questions itself?

No. SaveSlot keeps state; your own agent does the talking. Connect Claude, ChatGPT, Claude Code, Codex, Gemini, Grok or any remote-MCP assistant. See Connect an agent.

How is this different from my assistant's memory?

Memory keeps what was said. SaveSlot keeps what is currently true, as far as anyone knows: each value has a source, a version history and a trust marker, and conflicts are shown instead of quietly resolved. They work well together.

Can several agents share the same data?

Yes. A change one agent writes is what the next agent reads, with where it came from. If two agents write values that disagree, the record is marked as a conflict until you decide.

What if my agent guesses a number?

It should write it as Inferred. Other agents then read it as an estimate, not a fact. You can confirm or correct it, which adds a version. See Trust, versions and conflicts.

Do I need a data source?

No. Your agent can log things from conversation. Data sources add measured values so your agent does not depend on what you remember to mention.

Can I self-host?

Yes. The core is open source under Apache-2.0: see GitHub. Assistants that connect over the internet, such as Claude on the web or ChatGPT, need a server they can reach.

What happens to my data if I cancel Pro?

Your data is not deleted. Pro stays active until the end of the paid period; after that the Free limits apply, and everything comes back if you upgrade again.

How do I get help?

Write to support@saveslot.ai.