Audience
Developers and technical teams building or using AI agents that need persistent, source-linked context across work tools and AI sessions.
About Archilas
Archilas is a hosted memory service for AI agents that collects context from connected work tools and AI chats, then returns concise answers with named sources through MCP or its HTTP API. It imports connected-source history and continues syncing, organizes information into dated topic notes, and records who said or decided something and where it came from. When a newer decision replaces an older one, Archilas retains the earlier note as replaced so agents can use the current context while users can trace its history. Supported source connectors include Slack, GitHub, Gmail, Notion, Linear, Google Calendar, and Google Drive. Teams can use Archilas with MCP-compatible clients such as Cursor, Claude Code, ChatGPT, and Codex. The hosted service is aimed at developers and technical teams coordinating decisions across tools and AI sessions; plans vary by monthly question allowance and seats, and include a free tier.