Keep agent state in init.d-style files to survive compaction

Goal: Stop long agent sessions from losing their place after context compaction.
Setup: agents used in source: Claude, Cursor, Gemini CLI, Antigravity.
Steps:
1. Don’t trust the auto-summary after compaction; keep state on disk.
2. Avoid a single TODO.md/PROGRESS.md scratchpad; it gets stale or overwritten.
3. Use ordered, lexically numbered state files per step, like SysV /etc/init.d runlevels, that the agent re-reads to resume.
4. Push heavy work into forked subagents and wait on results, so the main context stays small.
Rules: ask me before anything irreversible (sending, posting, paying, deleting); don’t invent data, results or tools I haven’t given you; tell me what you couldn’t verify.
Based on: Randal L. Schwartz — https://dev.to/gde/surviving-the-200k-token-lobotomy-how-unix-initd-and-memento-made-my-ai-coding-agent-immune-to-2f74 (summarised from source; the source has no explicit step list)

Steps are our summary of the source (it has no explicit step list). Includes a “Based on” credit line.