feat: Claude Code Monitor — lanes, pipelines and a merged workspace

Internal SmartGift build of a Claude Code monitoring dashboard. Lanes: a durable unit of parallel agent work, one per working directory, tracked across session restarts. Managed lanes are git worktrees the dashboard provisions and can reset or remove behind a three-check destroy guard and a counted preflight; adopted lanes are directories you already own and are never destroyable. Pipelines: a lane moves through pipeline stages. A stage the agent declares with evidence renders green; a stage inferred from the tool-event stream renders dashed amber and never counts as done. Detection is forward-only within a 30-minute window, and never writes the declared stage. Workspace: one page at /run with a lane grid, the selected lane's pipeline, and a full Claude console behind a disclosure.
2026-07-29 17:07:45 +07:00
commit 57dc91585d
783 changed files with 221743 additions and 0 deletions
@@ -0,0 +1,65 @@
+---
+description: >
+  Discover when you are most active and most productive with Claude Code by
+  bucketing sessions and events into hour-of-day and day-of-week bins from their
+  timestamps, then flagging peak versus low-output windows. Uses the session
+  list, per-session events, and analytics daily trends. Use when planning a
+  schedule or deciding when to do deep work versus lighter tasks.
+---
+
+# Time of Day
+
+Profile activity and productivity across the hours of the day and days of the week.
+
+## Input
+
+The user provides: **$ARGUMENTS**
+
+This may be:
+- empty or "all" (default: all available sessions)
+- a window like "last 30 days" or "last 90 days" to limit the analysis
+- a project path to scope the analysis to one `cwd`
+
+## Data Sources
+
+| Endpoint | Returns |
+|----------|---------|
+| `GET /api/sessions?limit=500` | Sessions with `started_at`, `ended_at`, `status`, `cwd`, `cost`, and `metadata` (turn_count, total_turn_duration_ms) — primary source for hour/weekday bucketing |
+| `GET /api/events?session_id=X` | Events with `timestamp` and `event_type` (PreToolUse, PostToolUse, Stop, Compaction, APIError, etc.) — finer-grained activity within sessions and error timing |
+| `GET /api/analytics` | `daily_sessions` / `daily_events` (365d) and `sessions_by_status` for trend context and completion baselines |
+
+## Report Sections
+
+### 1. Activity by Hour of Day
+Bucket sessions (by `started_at`) and events (by `timestamp`) into 24 hourly bins.
+Show a text bar chart of session and event counts per hour. Identify the busiest
+hours by raw volume.
+
+### 2. Productivity by Hour of Day
+For each hour bin, compute completion rate (`completed / total` sessions started in
+that hour) and average sustained turn time
+(`total_turn_duration_ms / turn_count`, ms → minutes). Distinguish "active" hours
+(high volume) from "productive" hours (high completion + sustained turns).
+
+### 3. Day-of-Week Pattern
+Bucket the same metrics into 7 weekday bins. Table: weekday, sessions, completion
+rate, avg cost, dominant model.
+
+### 4. Peak vs. Low-Output Windows
+- **Peak windows:** hours/days with high completion rate and long sustained turns.
+- **Low-output windows:** hours/days with high abandonment/error/Compaction rates
+  or fragmented short turns. Pull error timing from `/api/events` event types
+  (APIError, Compaction) to corroborate.
+
+### 5. Schedule Recommendation
+Suggest which hour/weekday blocks to reserve for deep work and which to use for
+lighter or shallower tasks, grounded in the buckets above.
+
+## Output
+
+- Markdown with text-based bar charts (e.g., `09:00 ████████ 24`) for the hourly
+  and weekday distributions.
+- Tables for the hour and weekday metrics; ▲ / ▼ for above/below the overall mean.
+- Currency in USD to 4 decimals; durations in minutes (convert from ms).
+- Cite only numbers from the API. State how many sessions/events were bucketed and
+  exclude sessions missing `started_at` or the focus metadata, noting the count.