tasks ready

.opencode/skills/semantics-contracts/SKILL.md

@@ -33,8 +33,9 @@ Decision memory prevents architectural drift. It records the *Decision Space* (W
 2. **Task Guardrails:** Preventative `@REJECTED` tags injected by the Orchestrator to keep you away from known LLM pitfalls.
 3. **Reactive Micro-ADR (Your Responsibility):** If you encounter a runtime failure, use `explore()`, and invent a valid workaround, you MUST ascend to the `[DEF]` header and document it via `@RATIONALE [Why]` and `@REJECTED [The failing path]` BEFORE closing the task.
 
-**⚠️ `@RATIONALE`/`@REJECTED` ARE C5-ONLY.**
-Decision Memory tags belong exclusively to C5 contracts per Std:Semantics:Core complexity scale. C4 adds `@PRE`/`@POST`/`@SIDE_EFFECT` — not decision memory. Adding them below C5 violates INV_7 (verbosity/erosion). If a C1-C4 contract genuinely needs decision memory, it should be C5.
+**`@RATIONALE` / `@REJECTED` are ORTHOGONAL tags.** Per `axiom_config.yaml`, these are `protected: true` and `orthogonal: true` — they may appear at ANY complexity level (C1-C5) when a node records a deliberate architectural choice. They are REQUIRED for `ADR` type contracts. Removal of an existing `@RATIONALE`/`@REJECTED` requires `<ESCALATION>` to the Architect.
+
+If a C1-C4 contract records a workaround after a runtime failure, add `@RATIONALE`/`@REJECTED` at that node's header BEFORE closing the task. This is a Reactive Micro-ADR — it does NOT require bumping the complexity to C5.
 
 **Resurrection Ban:** Silently reintroducing a coding pattern, library, or logic flow previously marked as `@REJECTED` is classified as a fatal regression. If the rejected path is now required, emit `<ESCALATION>` to the Architect.