alive:session-context-rebuild

Session Context Rebuild

You're returning to work that spans multiple sessions — possibly conflicting ones. The context window is empty but the walnut isn't. This skill dispatches subagents to deeply reconstruct what happened, what changed, where conflicts exist, and what the merged state should be. Then you resume with full context as if you never left.

---

When to Use

• Returning to a walnut after days/weeks away
• Context was lost to compaction mid-session
• Multiple sessions touched the same walnut in parallel and need merging
• Picking up someone else's work in a shared walnut
• "What was I doing here?" moments

---

Process

Phase 1: Scout (single agent)

Dispatch one general-purpose subagent to map the territory. It reads the session-history/SKILL.md file and applies its browse logic against .alive/_squirrels/, then reports back:

• Which sessions touched this walnut, when, by whom (engine)
• Which files each session worked on (working: field in squirrel YAML)
• Which bundles were active
• Whether sessions overlap (parallel/conflicting work)
• Unsaved stash items from crashed sessions

Cross-walnut detection: the scout performs two queries, not one:

1. Walnut-tagged match — squirrel YAMLs where walnut: {target} (the obvious set)
2. Routed-stash match — squirrel YAMLs where any stash item's routed: field contains the target walnut name (catches maintenance sessions with walnut: null that routed work to this walnut)

Merge-deduplicate by session_id. Report both counts so the user sees "5 walnut-tagged + 1 cross-walnut routed" instead of just "5 sessions."

╭─ 🐿️ context rebuild — scouting
│  Found 5 sessions on this walnut (last 2 weeks)
│  3 sequential, 2 overlapping (possible conflicts)
│  12 files touched, 2 bundles active
│
│  ▸ Rebuild scope?
│  1. All 5 sessions (full rebuild)
│  2. Last 3 only
│  3. Just the 2 conflicting sessions
╰─

Phase 2: Deep Read (agent swarm)

Based on scope, dispatch parallel subagents — one per session or one per domain: