balanced7
UserMemory for AI agents that learns what actually helped: lessons injected at the moment of action, credited by outcome, challenged by counter-evidence. Local-first, multi-agent, test-gated.
Categories
Indexed Skills (3)
akashic-memory
Use whenever you work in the Akashic Aurora repo (E:\AI-Setup) — at session start, after ANY fix that first failed, whenever the user corrects you, and at session end. This is the shared-memory door; capture is the product, so use it even when the task feels too small to record. Also use when deciding whether knowledge belongs in a lesson, a note, a doc, or a hook.
root-cause-before-fixes
Use when encountering any bug, test failure, hook misbehavior, or unexpected output — BEFORE proposing or applying any fix, even a one-liner that looks obvious. Also use when a fix you tried didn't work, and especially after a third failed attempt.
verified-done
Use when about to claim any work is complete, fixed, passing, or shipped — before saying "done", "should work now", "that fixes it", or committing. Also use at the end of every slice, before wrap/handoff, and whenever you notice you are asserting an outcome you have not freshly observed.
Bio shown is the top-scored skill's repo description as a fallback — real GitHub bios land in a future update.