AI has no memory between sessions by default — so you re-explain everything. This skill closes the loop: Claude reads its own memory, your git history, and your notes, then hands you a military-style SITREP with paste-ready resume prompts.
Every session starts from zero. You spend the first ten minutes re-explaining your stack, your projects, and your last decision. That re-briefing ritual — not the model — is the real bottleneck.
One command triggers a full intelligence-gather before Claude writes a word. It reads its own memory files, yesterday's git commits, your wiki log, and your spec files — all in parallel.
Claude reads its own receipts so you never hand them over.
Active fronts (where each project stands now), yesterday's work (from real git commits, nothing invented), and open threads (ranked by how close to done). It reads like a staff officer briefed the commander.
The brief closes with a paste-ready prompt for every open thread. Copy it, open a new session, paste — and you're working in 30 seconds with full context. With great context comes great output.
A fill-in-the-blank SKILL.md that gathers memory + git + notes and outputs a morning SITREP with resume prompts.