28 lines
1.9 KiB
Markdown
28 lines
1.9 KiB
Markdown
# Bugs
|
|
|
|
# M1 — One agent end to end
|
|
|
|
# M2 — Seeing the work
|
|
|
|
# M3 — Second vendor and review
|
|
|
|
# M4 — Merging and history
|
|
|
|
# M5 — Workflows
|
|
|
|
- Workflow tab
|
|
A read-only right-pane tab on tasks with a `workflow_run`, showing the frozen graph (lazy-loaded Mermaid) and a textual stage list linking to each stage's task attempts and their finish_stage output. Each stage shows waiting on named dependencies, running, idle but not finished, finished, discarded before finishing, or discarded by normal cleanup after finishing, and which fork attempt fed the next stage. Backed by a typed run-progress API carrying the frozen definition, attempts, completion and dependency provenance, updated from sibling-task events.
|
|
- Start a workflow from the switcher
|
|
A "Start workflow…" entry beside "New task in …" in the switcher starts a saved workflow.
|
|
|
|
# M6 — Fan-out
|
|
|
|
# M7 — Inbox triage
|
|
|
|
- Human-facing task state
|
|
Switcher and Project rows show one derived label instead of the raw session state, by precedence: "Needs your answer" (an open inbox item), "Merging" or "Merge held", "Error", "Working", "Idle".
|
|
- Unread summary marker
|
|
Switcher and Project rows show a "New summary" marker, with text and an accessible label, when a task has a summary the user hasn't read. Selecting it opens the Transcript at that summary. Read state is per browser and per user, by event seq; a summary counts as read only when shown in a visible transcript or through "Mark seen", not by opening another tab or switching tasks. Task carries its latest summary seq so the marker survives reload without loading transcripts.
|
|
- Answer refinements
|
|
Choice blocks get clearer focus and arrow-key navigation. A "Let the agent decide" answer sends "Use your judgment for this decision and continue within the existing task instructions." and appears in the answer history; it never stands in for a required owner authorization.
|