The complete behavior contract, split by topic. The repository README explains why to use Wrighty, highlights the highest-impact workflows, compares the backends, and provides the first-run guide. These pages are the authority for options, guarantees, and edge cases.
- Configuration — backend selection,
wrighty init,.wrighty.json, validation. - User settings — installation-local profiles, model discovery, and the privacy-safe host label.
- Work items — IDs, creation and retry safety, editing, moving, archiving, custom fields, and import.
- Claims and ownership — claimant attribution, the ownership rules, fencing guarantees per backend, recovery paths, and escape hatches.
- Operator actions by surface — task-oriented comparison of what the Local web console, GitHub, and CLI can view or perform, with links to the authoritative procedures.
- Supported agents and surfaces — which agent families Wrighty supports for skills, headless work, session resume, Desktop opening, and handoff.
- Autonomous worker mode — eligibility, workspace modes, needs-attention and requeue, session resume, and the verified vendor capability matrix.
- Execution profiles — choosing a model and reasoning effort per run without work items naming a vendor model.
- Usage recovery and agent handoff — what happens when a subscription runs out mid-task: failure classification, deferred retry, the provider circuit, and opt-in cross-agent handoff.
- Web console — shared configuration and operations for both backends, plus the Local Markdown board/editor.
- Agent skills — installing and updating the bundled skill per agent surface.
- Storage and version control — what each backend stores where, and what to commit.
Related: workflow guide, item metadata, and design documents.