design_workflow_instructions: |
  For this repository, Jira project `PM` is both the human source of truth and the agent working truth for implementation tasks when Jira tools are available and permitted.

  Use Jira issue status, descriptions, comments, and linked Confluence pages as durable task state. Use Confluence space `PM` for larger design notes, product narrative, product decisions, and workplace-facing explanation. Local Markdown is supporting material only unless a task explicitly says otherwise.

  Polymodel is an AT Protocol-native model/project sharing app — roughly Thingiverse on atproto — being built for release as a real product. The tentative product shape is project/profile/like lexicons, profile/recent/hot feeds, and an in-browser STL viewer. `three-d` and `kiss3d` are candidate viewer libraries; prefer a spike before committing.

plan_handoff_instructions: |
  For Jira-backed tasks, cite the Jira issue key, current status, claim state, relevant Confluence links, jj workspace/bookmark name, validation commands, and review-gate requirements. The handoff should name `jira-solo-tasker-execute` when produced by the Jira solo-tasker workflow.

  Implementation plans should focus on building the atproto model-sharing app as a polished, releasable product.

orual_workflow:
  source_of_truth:
    human: jira
    agent: jira
  jira:
    cloud_id: "eab0a222-1a38-4539-a21d-44fb3466b28b"
    project_key: "PM"
    statuses:
      todo: "To Do"
      in_progress: "In Progress"
      in_review: "In Review"
      done: "Done"
    transitions:
      todo: "11"
      in_progress: "21"
      in_review: "31"
      done: "41"
    eligible_issue_types: [Task, Bug]
    automation_block_labels: [no-auto]
  confluence:
    space_key: "PM"
    space_id: "98315"
    overview_page_id: "98479"
    design_page_parent: null
  vcs:
    kind: jj
    workspace_policy: sibling
    workspace_command: "just workspace-create <name>  # creates sibling workspace and immediately starts a fresh jj change"
  review:
    gate_severities: [Critical, High]
    reviewer_models:
      - codex/gpt-5.6-sol(medium)
      - codex/gpt-5.6-luna(high)
      - zai/glm-5.2(max)