Skip to Content
ReferenceGlossary

Glossary

Core terms used across AgentTeams.

TermDescription
PlanA unit that defines the intent and scope of work. The agent executes it once a human approves.
AgentAn AI agent that performs the work (e.g. claude-main). You can run several, organized by role.
RunnerA local executor that picks up requests and runs the configured agent. Registers as an OS service and starts automatically.
RunnerBoxThe isolated workspace a runner uses to carry out work.
ConventionProject rules/guides an agent follows when writing code or doing work. Organized into rules, skills, guides, and references categories.
CoActionA handoff record that transfers work state and context between sessions. Its source is either Runner (auto-created by a runner session) or Direct (created manually).
Completion ReportA report capturing a finished task’s summary, verification results, and quality score.
Post-mortemA document analyzing the root cause of a failure and its prevention measures, which are linked to Conventions.
Code ReviewA workflow that verifies changes independently of plan completion.
DocumentA project document for human-facing artifacts (design notes, onboarding, etc.).
Last updated on