Reference applications built with Awaken Agents.

01 · Our reference application

Awaken Pilot

Create a thin Agent application

Source-archive reference build

A task interface for progress, approvals and deliverables.

The situation: You have the product idea, but the first prototype still needs task state, progress, approval, schedules, files, and recovery.

Start with: Run one bounded Mission in Plan or Execute mode, then review its progress and result in the same Session.

Best for: Agent product teams that want the shortest path from product UX to durable execution.

See the Pilot reference build

02 · Independent reference build

DeepSeek workbench on Awaken Agents

Build an advanced Agent workbench

Reference build with fixture and live browser evidence

Keep the workbench experience and use Agents to run its tasks.

The situation: An interaction-rich workbench needs Presets, Plans, approvals, Trajectory, Threads, and Files. Rebuilding the Agent loop first delays the parts users can actually see and judge.

Start with: Start one Preset-driven Session, review its Plan, approve a tool, and reopen the same Trajectory after an interruption.

Best for: Teams building Agent IDEs, coding workbenches, or other interaction-rich Agent products.

See the reference reimplementation

03 · Our reference application

Awaken Design

Build a domain-rich AI product

Locally qualified reference build

Turn one AI-generated design into a reviewable, revisable, explicitly accepted result.

The situation: Chat history is not a product model when a team must compare versions, attach feedback to the exact result, reject unsafe output, and deliver one selected design.

Start with: Submit a design brief, review the generated version, request a change, then select and export the design to deliver.

Best for: Development teams turning AI-generated results into products people can revise, review and receive.

See the Awaken Design reference build