https://www.duncanleung.com/blog 📚 Learning, 👨🏻‍💻 Building, 📝 Documenting... en-us duncan@leungs.email (Duncan Leung) duncan@leungs.email (Duncan Leung) Tue, 22 Sep 2026 00:00:00 GMT https://www.duncanleung.com/blog/claude-code-precompact-postcompact-context-management https://www.duncanleung.com/blog/claude-code-precompact-postcompact-context-management Claude Code's /compact command compresses your conversation to free context, but it silently drops external state mutations, abandoned approaches, and behavioral constraints. I built a precompact/postcompact checkpoint system that writes ground truth to disk before compaction and cross-checks it afterward — including live reconciliation against git, CI, and external systems. Here's the design, the research behind it, the failure modes it prevents, and how OpenRig's native compaction system eventually superseded these custom skills. Tue, 22 Sep 2026 00:00:00 GMT duncan@leungs.email (Duncan Leung) aiclaude-codeopenrig https://www.duncanleung.com/blog/openrig-compaction-system-why-compact-instead-of-handoff https://www.duncanleung.com/blog/openrig-compaction-system-why-compact-instead-of-handoff When a long-running orchestrator hits context limits, the obvious move is to hand off to a fresh agent. We tried that. The startup overhead, lost momentum, and broken continuity made it worse than the problem it solved — for our use case. So we built a compaction system instead: one that writes ground truth to disk, restores from it, and distills session knowledge into a wiki page via an API call to Haiku. Recent research shows neither approach is categorically better — compaction drops behavioral constraints, handoff carries a 'handoff tax' — but for same-model orchestrators that accumulate judgment over time, compaction with external ground truth wins. Here's how the system works and what the literature says. Mon, 19 Oct 2026 00:00:00 GMT duncan@leungs.email (Duncan Leung) aiclaude-codeopenrigmulti-agent