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-quality-technical-debt/SKILL.md

@@ -0,0 +1,63 @@
+---
+name: evolv-quality-technical-debt
+description: Drive code quality, regression prevention, and technical debt management for EVOLV nodes. Use when reviewing changes for bugs, maintainability, test completeness, architectural drift, and when creating prioritized remediation plans for JavaScript/CommonJS Node-RED modules.
+---
+
+# EVOLV Quality Technical Debt
+
+## Mission
+Raise delivery reliability by detecting defects early and systematically reducing technical debt in EVOLV nodes.
+
+## Harness Execution Contract
+- Anchor findings to concrete file/line evidence.
+- Separate correctness risk from style preferences.
+- Require regression-proof evidence for fixes (tests that fail-before/pass-after when feasible).
+- Feed recurring failure patterns back into the relevant skill guidance.
+
+## Scope
+- Node implementation quality in `nodes/<nodeName>/src/`
+- Editor/runtime contract consistency in `.html` + runtime wrappers
+- Shared utility hygiene in `nodes/generalFunctions/`
+- Test depth in `nodes/<nodeName>/test/`
+
+## Test Policy Baseline
+All code changes require tests under `nodes/<nodeName>/test/`.
+
+Cover at minimum:
+- config default/override behavior and numeric coercion
+- each supported `msg.topic` handler with edge cases
+- child registration and dedupe side effects
+- tick/output boundaries and error paths
+- regression tests for fixed bugs
+
+Execution:
+- preferred: `node --test nodes/<nodeName>/test/*.test.js`
+- fallback: `node nodes/<nodeName>/test/<file>.test.js`
+
+## Review Workflow
+1. Assess correctness risks first (runtime errors, logic regressions, broken topic contracts).
+2. Assess maintainability (duplication, unclear ownership, implicit behavior).
+3. Assess test adequacy against EVOLV policy:
+- config defaults/overrides
+- topic handlers and edge cases
+- tick/output boundaries
+- regressions for fixed bugs
+4. Create a prioritized debt backlog with effort and impact.
+
+## Quality Criteria
+- Domain logic remains testable without full Node-RED runtime.
+- Complex logic is explicit and minimally coupled.
+- Backward compatibility is deliberate and documented.
+- New behavior includes tests that fail before and pass after.
+
+## Deliverables
+Return:
+- findings by severity
+- test gaps and specific cases to add
+- debt backlog (now/next/later)
+- recommended refactors with expected payoff
+
+Decision interview triggers:
+- tradeoff between delivery speed and known high-severity risk
+- acceptance of temporary risk with deferred remediation
+- testing scope reductions that materially raise regression risk