Back to directory

Promptbook: Invisible AI Agents

A platform for defining persistent AI agents in a plain-language format called Books, then deploying them to an Agents Server where they run continuously, remember context and work in teams.

Promptbook: Invisible AI Agents

Promptbook lets you define AI agents using "Books", a human-readable, explicit language for describing an agent's goal, rules and knowledge sources, meant to be easier to write and maintain than raw prompts.

Agents are deployed to the Agents Server, where they run persistently, remember context, collaborate with other agents, and carry out ongoing goals rather than answering single requests. A standalone installer sets up the server on a fresh VPS in interactive or non-interactive mode, with environment presets for preview, production or long-term-support deployments; providing an OpenAI API key configures the default OpenAI Codex runner automatically, and other model runners can be configured afterward from the UI or over SSH.

It is written in TypeScript.

More in Agent frameworks

See all 902
  • Recurse

    recurse.run

    A serverless harness that lets a coding agent design, test, and deploy its own specialist sub-agents as tools, MCP servers, or bots, defined by a schema-validated manifest.

  • ProofForge

    shulgin.is-a.dev

    An AI-agent pipeline that decomposes a mathematics problem, proves each piece, and formalizes the result in Lean 4 and Mathlib until the Lean kernel accepts it, with six of its proofs already merged into Google DeepMind's formal-conjectures repository.

  • local-coder

    github.com

    Builds a local coding-agent team on top of Ollama and OpenCode by detecting the machine's hardware, recommending models for seven specialist roles, and testing whether each assigned model can actually use tools, edit a repository and pass verification.