Six pillars for agents that take action.
Munin reads every prior Claude Code, Codex, OMX, and raw PowerShell session across all projects, all time. It journals, deduplicates, and reconciles every command, correction, redirect, and verification into one cross-tool history.
Munin builds the memory from what already happened.
Import your strategy document: goals, KPIs, initiatives, constraints, assumptions. Munin compiles it into a deterministic kernel and joins it with your evidence corpus. Each nudge carries the task, why it matters now, supporting evidence, confidence, and expected effect.
The next strategic task, tied to your priority.
Action candidates earn their way to rules. Munin ranks by precedent count, success count, and failure count across every session. Candidates promote to assertions, assertions promote to rules, observe-only by default.
Every promoted claim carries freshness, stability, and confidence.
prisma.migration → prisma.db.pushpnpm testpnpm run lint cleanprisma db push for local schema syncpnpm test before commit on sitesortedMunin maintains a per-project kernel: open loops, live claims, checkpoints, current recommendation, active risks. Every session re-entry gets a first question to answer and a first verification to run.
The agent hits the right target because the target is explicit.
Munin reports repeated corrections, redirect statistics, likely misunderstandings, and behaviour-change recommendations per agent. The same memory surface tells Claude Code, Codex, and IDE agents what to change. Leaves the others alone.
One surface, distinct directives per agent.
Resume and handoff briefs stay active while the latest replay/eval against a held-out corpus verifies. If proof drops, Munin falls back to a deterministic packet path.
No silent regressions. No trust-me-bro memory.
.prettierrc path in agent preamblex-munin-sig rule to memorytsc --noEmit to pre-commit gate