Glossary

Work and projects

• Project key. The short identifier for a project, used to distinguish its work. Choose it when you create a project.
• Work item. An Epic, Feature, Story, Bug, or Task. Work items have a type-specific state and belong to a project. See work items.
• State. A label that tracks a work item’s place in its workflow. The default labels are New, Active, Resolved, Closed, and Removed. Projects can configure their own states. See work item states and board settings.
• Category. The meaning assigned to a state: Proposed, In Progress, Resolved, Completed, or Removed. Boards show states. ShipLean uses categories to understand what each state means. See board settings.
• Epic. A broad area of work that can contain Features. See Epics.
• Feature. A piece of work that can contain other Features, Stories, and Bugs. It can hold a goal and the research record. See Features.
• Story. A work item that describes a deliverable and can contain Tasks. Story points are estimates on Stories and Bugs. See work items.
• Bug. A work item for a defect. Bugs can contain Tasks and can have story point estimates. See work items.
• Task. A work item for a piece of work under a Story or Bug. See work items.

Planning

• Sprint. A time-boxed period for planned work. Sprints have dates and capacity. See sprints.
• Capacity. The amount of work a person or team can take into a sprint, measured in story points. See planning and capacity.
• Velocity. The story points completed per sprint, used to plan future work. See planning and capacity.
• Forecast. An estimate of how much planned work fits in a sprint, based on its capacity and the selected velocity. See planning and capacity.
• Swimlane. A horizontal grouping of cards on a board. For example, a Features board can group work by Epic. See boards.
• WIP limit. A maximum number of cards allowed in a board column. See board settings.

Agents and access

• Agent. An MCP client connected to ShipLean as its own user. A person approves its requested access and can later revoke the connection. See connecting an agent.
• Scope. A named permission an agent requests, such as reading or writing project work. Review scopes before you approve access. See agent scopes.
• Consent. The approval step where you review an agent’s identity and requested scopes, then allow or deny the connection. See agent consent.