Two board-level additions:
Merge all reviewable agent branches into the main branch sequentially:
- worktrees.mergeBranches() merges candidates in order, advancing HEAD each
time; refuses if the main tree is dirty, and on the first conflict aborts
that merge (main left clean) and stops — prior merges stay, untried
branches reported as unattempted.
- runManager.mergeAll() picks candidates (settled runs that own their
worktree, branch present, not yet in HEAD; deduped, oldest-first) — so
refinement runs (which inherit a worktree) are correctly excluded.
- POST /runs/merge-all; ⬇ Merge all button in the board header opens a
results modal (per-branch ✓/⇢/⚠, conflict banner, inline git output).
Floating dock to view/open running agents (front-end only):
- RunningAgentsBar pins bottom-right (below the card modal), lists every
running run with live elapsed times, and opens that card's modal on click;
auto-hides when nothing runs, collapsible otherwise.
EOF && echo "" && git log --oneline -4
Two intertwined changes that both touch the orchestrator hook + run console:
Isolate the agent event stream (perf):
- useRunStream owns the SSE stream + event log locally inside AgentRunBar, so a
burst of streamed events re-renders only the console — never the board page or
card modal (which was causing frame drops at run start).
- useOrchestrator is now a registry only; lifecycle events reflect back up via
stable patchRun/reflectBevy reflectors (effect deps depend on those, not the
whole object, avoiding a stream-teardown loop).
Session resume for Refine:
- Runs now persist their pi session (drop --no-session); each fresh run captures
its session JSONL path into a new agent_runs.session_file column (additive,
idempotent migration).
- Refine resumes the prior run's actual session (--session <path> → appends) in
that run's own worktree (inherited, never owned), sending the operator's
feedback as the next message in the same conversation with full prior context.
- owns_worktree guards remove()/cleanup so a refinement never destroys the
owning run's worktree; bad refinement targets return 409.
- AgentRunBar shows Refine only for settled runs with a recorded session.
EOF && echo "" && git log --oneline -3
- RunEventList: grouped activity timeline. Assistant text becomes chat
bubbles (auto-collapsing long messages); tool_start/tool_end pair into
entries with spinners and expandable input/result blocks; bevy output
rolls into a live console; relative timestamps on a left rail
- AgentRunBar: redesigned as a mission console. Live stats header (elapsed
time, tool count, events), animated status banner with sweep/glow while
running, clearer action bar. All controls preserved (run/steer/stop,
diff/merge/bevy) so the human-only merge/complete safety model holds
- tailwind.css: vn-flow, vn-sweep, vn-dots, vn-spin keyframes
- CardModal: full card overlay (orchestrator, references, tags, comments)
- DiffModal: branch-diff review (commits, stat, capped patch)
- useOrchestrator: background polling + bevy status sync + ref-counted SSE
- KanbanCard: pulsing agent/bevy running badge on collapsed cards
Replace the localStorage kanban with the backend-backed board, add typed clients
and a React hook with optimistic updates. Cards can reference static doc pages
and user-created custom pages (new "custom" reference type with purple chips).
Add the agentic orchestrator UI: a per-card panel to launch `pi` runs, watch a
live tool/thought stream over SSE, steer mid-run, and stop — while the board
stays fully interactive. The board page wires the orchestrator and custom-pages
stores into every card.
