Origami
A development workspace orchestration tool built by an individual developer that integrates terminals, AI agents, browser tabs, Git, and kanban boards into a unified workspace. It does not run models or host keys, positioning itself as a disposable, lock-in-free orchestration layer.
Timeline (last 90 days)
Origami is positioned as a workspace that integrates development tools and AI agents; it does not run models or host users' API Keys, but instead drives agents that are already installed and logged in.
Origami adopts a 'project = workspace' model, where a single workspace simultaneously holds terminal sessions, running AI agents, browser tabs, Git operations, kanban cards and documents produced by agents.
Origami integrates MCP (Model Context Protocol), enabling agents to run commands and read output, drive browser tabs, create cards on the kanban board, and comment on code diffs.
Origami explicitly promises not to run models, not to hold keys, and to follow users' existing login methods; the author says that if users stop using it tomorrow, their development environment will not change at all.
Origami is currently at v27, and the author is soliciting feedback from the community.
Origami is a personal project.
In Origami, an 'agent' refers to an AI program with autonomous action capability that can break down tasks according to goals, call tools, execute operations, and adjust subsequent steps based on results.
The cost of Origami's lock-in-free design is that it cannot provide the out-of-the-box experience that deep binding brings; users need to maintain the underlying agents and toolchain themselves.
Considering that Origami is a personal project, its ecosystem maturity and long-term maintenance capability remain to be seen.
All Facts (9)
Origami is positioned as a workspace that integrates development tools and AI agents; it does not run models or host users' API Keys, but instead drives agents that are already installed and logged in.
50%UnverifiedOrigami adopts a 'project = workspace' model, where a single workspace simultaneously holds terminal sessions, running AI agents, browser tabs, Git operations, kanban cards and documents produced by agents.
50%UnverifiedOrigami integrates MCP (Model Context Protocol), enabling agents to run commands and read output, drive browser tabs, create cards on the kanban board, and comment on code diffs.
50%UnverifiedOrigami explicitly promises not to run models, not to hold keys, and to follow users' existing login methods; the author says that if users stop using it tomorrow, their development environment will not change at all.
50%UnverifiedOrigami is currently at v27, and the author is soliciting feedback from the community.
50%UnverifiedOrigami is a personal project.
50%UnverifiedIn Origami, an 'agent' refers to an AI program with autonomous action capability that can break down tasks according to goals, call tools, execute operations, and adjust subsequent steps based on results.
50%UnverifiedThe cost of Origami's lock-in-free design is that it cannot provide the out-of-the-box experience that deep binding brings; users need to maintain the underlying agents and toolchain themselves.
50%UnverifiedConsidering that Origami is a personal project, its ecosystem maturity and long-term maintenance capability remain to be seen.
50%