Frequently Asked Questions

Does Lore work with ChatGPT?

Yes. Lore works with any MCP-compatible AI client. This includes Claude, ChatGPT (on Business, Enterprise, and Edu plans), OpenClaw, and any other tool that supports the Model Context Protocol (MCP). The user connects their AI client to Lore by adding the Lore MCP URL (https://uselore.app/mcp) to the client's MCP server configuration and signing in to Lore when the client asks. The URL is the same for every account; the sign-in is what identifies the user.

Can I see what my AI has stored?

Yes. The Lore web interface at uselore.app shows all entries in the user's knowledge base with their full content, entry type, tags, status, description, and timestamps. Users can browse, read, edit, and archive any entry.

What happens if I switch AI providers?

The user's Lore knowledge base persists independently of any AI provider. Switching from one AI to another does not affect the knowledge base. The user connects the new AI client to the same MCP URL. All existing entries remain accessible.

Is my data private?

Yes. Each Lore user has an isolated SQLite database, encrypted at rest. User data is not shared between users. Lore does not train models on user data, and does not sell it. On keyword-only search nothing in a knowledge base is sent to any third party. On accounts with semantic search — the standard search since 15 August 2026 — section text is sent to VoyageAI to be turned into vectors when entries change, and the search query, the candidate sections and each candidate entry's name, type and description are sent when a search by meaning runs; nothing else leaves Lore, and keyword-only search is available on request. Lore does not run AI models — all AI processing happens on the user's chosen AI provider. See the Data and Trust page and Semantic Search and Your Data for full details.

Can I delete entries?

No. Entries cannot currently be deleted permanently. Users can archive an entry through the Lore web interface; archived entries stay in the user's database. Individual sections of an entry can be deleted.

What is MCP?

MCP stands for Model Context Protocol. MCP is an open standard that allows AI assistants to connect to external tools and data sources. When an AI client supports MCP, the AI can call tools provided by MCP servers during conversations. Lore is an MCP server for reading from and writing to a user's knowledge base: a new account sees 11 tools — 9 core tools, plus open_entry and review_pending_action from the MCP App, which is on for every account by default — and enabling comments adds list_comments and comment, for 13 in all. One more, resolve_pending_action, is app-only: the widget calls it and the AI never sees it. The MCP specification is maintained by Anthropic and is supported by a growing number of AI clients.

Do I need to be technical to use Lore?

No. The user adds one URL to their AI client's MCP configuration, signs in to Lore once, and has a conversation. The AI handles the knowledge base setup, entry creation, organisation, and retrieval. The user does not need to write code, manage databases, or understand technical concepts. The setup process is a guided conversation between the user and their AI assistant.

What does "AI writes, human reads" mean?

During normal conversations, the user's AI assistant creates structured knowledge base entries — decisions, insights, project specifications, research notes, meeting summaries. The user browses, reads, edits, and owns everything through the Lore web interface. The user does not have to write or organise anything manually. The AI does the structuring based on a personalised operating manual that defines entry types, content templates, tags, and naming conventions for each user.

How is Lore different from ChatGPT Memory or Claude Memory?

Platform memory features (ChatGPT Memory, Claude memory) store opaque facts that the user cannot fully see, structure, or take with them to another AI provider. Lore stores structured, readable, editable entries organised by type, tags, and status. Lore entries work across all MCP-compatible AI tools simultaneously. Lore entries belong to the user and are not locked to any AI platform.

Can I use Lore with multiple AI tools at the same time?

Yes. The user can connect Claude, ChatGPT, OpenClaw, and any other MCP-compatible AI client to the same Lore knowledge base simultaneously. All connected AI clients read from and write to the same knowledge base. An insight that Claude saves is immediately available to ChatGPT and any other connected AI.

How much does Lore cost?

Lore is currently in invite-only early access at no cost. Pricing will be published on the Pricing page when finalised. Because Lore does not run AI models, Lore has zero LLM costs — pricing will reflect infrastructure and development costs, not per-query AI charges.

What data does Lore store?

Lore stores the knowledge base entries that AI assistants create via MCP tool calls. Each entry has a name, type, content (Markdown), description, tags, status, source, and timestamps. Lore also stores the user's KB configuration (entry types, content templates, tag taxonomy, status lifecycle, naming conventions). Lore does not store conversation transcripts, AI chat history, or any data that the AI does not explicitly write to the knowledge base.