docs: retire repo-mem MCP, migrate skills to .claude/skills, audit fixes

- Delete .mcp.json + .claude/rules/repo-mem.md; drop .repo-mem from .gitignore
- Remove repo-mem / substrate_score / repo_search references from all .md
- Move 15 EVOLV skills from .agents/skills/ to .claude/skills/ so they are
  auto-discovered by the Claude Code harness and invokable via the Skill tool
- Retire .agents/skills/evolv-orchestrator (duplicate of the subagent at
  .claude/agents/evolv-orchestrator.md); orchestrator lives as a subagent only
- Drop OpenAI-format agent yaml metadata from each skill (not needed for CC)
- Update CLAUDE.md, CONTRACTS.md, AGENTS.md to point at the new locations and
  disambiguate skills (.claude/skills/) vs subagents (.claude/agents/)
- Fix CLAUDE.md tick-loop wording (opt-in per-node, not a fixed 1000ms)
- Widen .claude/rules/ paths frontmatter so node-architecture and telemetry
  rules trigger on more relevant files; add frontmatter to flow-layout rule
- Bump CONTRACTS.md review date to 2026-05-19; add step 7 to the contract-
  change workflow (review example flows when topic usage changes)
- Bump nodes/generalFunctions pin (Home.md substrate_score reference removed)

Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>

.claude/skills/evolv-alarms-interlocks-permissives/SKILL.md

@@ -0,0 +1,54 @@
+---
+name: evolv-alarms-interlocks-permissives
+description: Design and review alarms, interlocks, and permissive logic for EVOLV control nodes. Use when implementing trip conditions, permissive checks, startup/shutdown guards, alarm priorities, latching/reset behavior, and operator-facing fault handling.
+---
+
+# EVOLV Alarms Interlocks Permissives
+
+## Mission
+Make alarm and interlock behavior explicit, testable, and operationally safe while preserving availability-first policy bounds.
+
+## Harness Execution Contract
+- Build alarm/interlock map from current node contracts and state logic.
+- Define invariants before edits:
+  - trips/permissives are deterministic
+  - latching/reset behavior is explicit
+  - operator-visible diagnostics are preserved
+- Validate with sequence and fail-state tests.
+
+## Scope
+- `nodes/pumpingStation/`
+- `nodes/machineGroupControl/`
+- `nodes/rotatingMachine/`
+- Any node with mode/state transitions and protective actions
+
+## Workflow
+1. Enumerate alarm conditions and priority/severity.
+2. Define interlock and permissive truth tables.
+3. Verify startup/shutdown/emergency sequences.
+4. Confirm reset, auto-recovery, and manual acknowledgement behavior.
+5. Ensure outputs expose actionable fault context.
+
+## Standards
+- Avoid hidden permissives; every gate should be observable.
+- Keep alarm naming stable and semantically clear.
+- Separate advisory warnings from trip-level protection.
+- Preserve controlled compatibility for released fault topics.
+
+## Test Expectations
+Cover:
+- trip activation and reset/latch behavior
+- permissive-denied and permissive-restored transitions
+- out-of-order signal handling in sequence transitions
+- degraded sensor quality paths and alarm escalation
+
+## Deliverables
+Return:
+- alarm/interlock/permissive matrix
+- changed files/tests and evidence
+- unresolved protection-vs-availability tradeoffs
+
+Decision interview triggers:
+- changed trip thresholds or permissive logic with operational impact
+- altered reset authority (auto vs manual)
+- alarm contract changes affecting external consumers