Issues

• [program §8] Edge が所有する7つのサービス目標を測るハーネスを実装する

  01rabbit/Azazel-Edge#414 · Filter by author 01rabbit opened on Sep 19, 2026

• [R0/R3] Declare which decision/enforcement path is authoritative in the x86_64 integration boundary

  architecturearea:integration
  01rabbit/Azazel-Edge#412 · Filter by author 01rabbit opened on Sep 19, 2026
  1

• [roadmap] Outcome-as-Evidence → Pi/HIL → Active Defense differentiation execution sequence

  01rabbit/Azazel-Edge#408 · Filter by author 01rabbit opened on Aug 29, 2026
  3

• [implementation baseline] Outcome-as-Evidence v1 — shadow execution/mechanism/outcome loop without hot-path replacement

  architecture
  01rabbit/Azazel-Edge#400 · Filter by author 01rabbit opened on Aug 25, 2026
  9

• review: adversarial differentiation gate for Tempo/Terrain/Belief/Council/Co-Adaptation

  01rabbit/Azazel-Edge#399 · Filter by author 01rabbit opened on Aug 25, 2026

• mio: implement outcome-validated Operator Doctrine and Human-AI Team Memory

  01rabbit/Azazel-Edge#398 · Filter by author 01rabbit opened on Aug 25, 2026

• mio: implement independent strategic council with preserved dissent and no majority authority

  01rabbit/Azazel-Edge#397 · Filter by author 01rabbit opened on Aug 25, 2026

• mio: build counterfactual Defensive Effect planner for replay/shadow wargaming

  01rabbit/Azazel-Edge#396 · Filter by author 01rabbit opened on Aug 25, 2026

• mio: maintain evidence-bounded Adversary Belief State without attribution overclaim

  01rabbit/Azazel-Edge#395 · Filter by author 01rabbit opened on Aug 25, 2026

• architecture: model Real Terrain and Presented Terrain as separate provenance-bearing defensive state

  01rabbit/Azazel-Edge#394 · Filter by author 01rabbit opened on Aug 25, 2026

• research: define Adversary Tempo model and measurable Defensive Initiative metrics

  01rabbit/Azazel-Edge#393 · Filter by author 01rabbit opened on Aug 25, 2026

• architecture: define tool-agnostic Defensive Effect Protocol above concrete enforcement adapters

  01rabbit/Azazel-Edge#392 · Filter by author 01rabbit opened on Aug 25, 2026

• [epic] Next-Generation Azazel Active Defense Control Plane — Tempo, Terrain, Belief, Co-Adaptation

  01rabbit/Azazel-Edge#391 · Filter by author 01rabbit opened on Aug 25, 2026
  5

• architecture: migrate outward vocabulary to canonical Defensive State

  area:ai
  01rabbit/Azazel-Edge#390 · Filter by author 01rabbit opened on Aug 24, 2026

• mio: complete reasoning-trace lifecycle and minimized audit retention

  area:ai
  01rabbit/Azazel-Edge#389 · Filter by author 01rabbit opened on Aug 24, 2026

• mio: publish canonical cognitive-plane architecture and authority matrix

  area:ai
  01rabbit/Azazel-Edge#388 · Filter by author 01rabbit opened on Aug 24, 2026

• mio: complete grounding, playbook provenance, and architecture documentation

  area:ai
  01rabbit/Azazel-Edge#386 · Filter by author 01rabbit opened on Aug 24, 2026

• mio: complete reasoning lifecycle, supersession, and no-model handoff

  area:ai
  01rabbit/Azazel-Edge#385 · Filter by author 01rabbit opened on Aug 24, 2026

• mio: complete typed read-only Capability Broker contract and audit

  area:ai
  01rabbit/Azazel-Edge#384 · Filter by author 01rabbit opened on Aug 24, 2026

• mio: close SituationFrame contract and Pi-budget acceptance gaps

  area:ai
  01rabbit/Azazel-Edge#383 · Filter by author 01rabbit opened on Aug 24, 2026

• [architecture] Make Defensive State the canonical outward state vocabulary

  01rabbit/Azazel-Edge#379 · Filter by author 01rabbit opened on Aug 22, 2026
  1

• docs: finalize M.I.O. cognitive-plane architecture, operations, offline deployment, and field-validation package

  area:aiarea:integration
  01rabbit/Azazel-Edge#375 · Filter by author 01rabbit opened on Aug 22, 2026
  2

• mio: expose reasoning, hypotheses, provenance, and authority boundaries in SENTINEL operator surfaces

  area:aiarea:integration
  01rabbit/Azazel-Edge#374 · Filter by author 01rabbit opened on Aug 22, 2026
  2

• review: adversarial M.I.O. Cognitive Plane gate before cross-product reasoning integration

  area:aiarea:integration
  01rabbit/Azazel-Edge#373 · Filter by author 01rabbit opened on Aug 22, 2026
  1

• mio: build reproducible evaluation/replay harness for reasoning quality, safety, and Pi resource budgets

  area:ai
  01rabbit/Azazel-Edge#372 · Filter by author 01rabbit opened on Aug 22, 2026
  5