Work
Threads and projects
Keep everyday AI work close to a familiar conversation without losing the project it belongs to.
- Thread history
- Project instructions
- Files and context
The product
Work in familiar threads while the role, method, source context, run record, and reviewed output remain available underneath.
Invite-only preview. Public resources remain free.
Client workspace
Landing page audit / example run
Output reusable
Thread
Audit this landing page and prioritise the fixes for the next build pass.
Saved output
01Workspace model
The hierarchy is intentionally small. Each level answers where the context belongs and what should survive the session.
Workspace
Client, brand, or initiative
The private operating context for a body of work.
Project
Campaign or engagement
Instructions, files, threads, outputs, and decisions stay grouped.
Thread
Daily working surface
Start in plain language, then attach the right operating context.
Library
Reviewed working material
Keep the outputs worth reusing, exporting, or attaching next time.
02Capabilities
The interface stays calm because the structure appears where it is useful, not as setup work before every conversation.
Threads and projects
Keep everyday AI work close to a familiar conversation without losing the project it belongs to.
Operators, playbooks, knowledge
Attach the role, method, source context, and quality bar before a model starts guessing.
Run records and review
See the prompt snapshot, source context, validation state, trace, and review notes when trust matters.
Outputs and patterns
Promote reviewed work into a library instead of saving every answer or losing the useful ones.
03The loop
Request captured
The everyday surface stays calm: one request, the relevant project, and the conversation around it.
Active run
ThreadRequest
Audit this landing page.
Context
Operator + playbook + 3 sources
Output
Prioritised fix list
04Governance
Source context, prompt snapshots, validation, execution labels, and review decisions remain legible when the work needs trust.
Walk through a complete runChoose the next path
Newsletter
Practical prompts, operator ideas, playbook patterns, and product notes from building a private AI workspace for repeatable work.