webgptorg/promptbook

webgptorg/promptbook

Agent

Agent that update models and LLM providers.

webgptorg/promptbook

Agent

Every agent has a visual identity used across the UI (directory cards, chat header, manifest icons, social previews). It is resolved from the agent's book with deterministic fallbacks, so an agent always has an avatar.

webgptorg/promptbook

Agent

Federation lets one Agents Server instance see and reuse agents hosted on other Agents Servers. It is configuration-driven and read-only: a federated server is never written to; its agents are referenced by URL or copied on demand.

webgptorg/promptbook

Agent

Agents are organized in a folder tree per server instance. Folders group agents in the directory (homepage, header menu, dashboard) and drive bulk operations (visibility, export). State lives in prefixAgentFolder plus Agent.folderId/Agent.sortOrder (Data model).

webgptorg/promptbook

Agent

How one stored agent source becomes the effective (resolved) source that is actually compiled and executed. Resolution expands FROM (inheritance) and IMPORT (textual inclusion) commitments, rewrites compact agent references into canonical URLs, and degrades gracefully when references cannot be loaded.

webgptorg/promptbook

Agent

Compiling an agent (resolving inheritance, building model requirements, indexing KNOWLEDGE into provider-side vector stores) is expensive. The server therefore prepares agents in the background ahead of chat time and caches every derived artifact by content hash, so the first user message does not pay the indexing…

webgptorg/promptbook

Agent

Agents whose book is open learn from their conversations: after a turn, the executing agent may append new content to its own source (samples of the exchange, teacher-suggested commitments). The server persists these changes append-only, so learning can never rewrite or corrupt the authored book.

webgptorg/promptbook

Agent

How agents move between servers and how a server's state is exported. Everything is book-centric: because a book is the single source of truth for an agent's behavior, plain .book files are the portable format.