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-instrumentation-assets/SKILL.md

@@ -0,0 +1,55 @@
+---
+name: evolv-instrumentation-assets
+description: Engineer measurement and instrumentation behavior for EVOLV assets. Use when defining sensor mappings, tag semantics, data quality checks, measurement container behavior, calibration assumptions, and measurement-driven node logic in `measurement` and related assets.
+---
+
+# EVOLV Instrumentation Assets
+
+## Mission
+Ensure measurements are trustworthy, well-modeled, and usable by control and analytics logic.
+
+## Harness Execution Contract
+- Ground changes in current measurement definitions and downstream consumers.
+- Define invariants before edits:
+  - unit consistency and conversion transparency
+  - explicit quality-state handling
+  - no silent coercion that hides sensor faults
+- Provide evidence for data-quality transitions and fallback behavior.
+
+## Scope
+- `nodes/measurement/`
+- `nodes/generalFunctions/src/helper/measurement*`
+- `nodes/generalFunctions/datasets/assetData/measurement.json`
+- Any node consuming measurement topics or quality flags
+
+## Workflow
+1. Inventory required measurements by asset/function.
+2. Define tag semantics and expected units.
+3. Add validation rules for bounds, type, and missing values.
+4. Specify handling for stale/noisy/bad-quality signals.
+5. Ensure downstream nodes receive consistent payload structure.
+
+## Standards
+- Prefer explicit unit naming and conversion points.
+- Separate raw sensor value from engineered value when possible.
+- Avoid hidden coercions that mask instrumentation faults.
+- Record assumptions around calibration and sensor placement.
+
+## Test Expectations
+Cover:
+- missing/invalid payloads
+- unit conversion correctness
+- quality/state transitions (good/suspect/bad)
+- behavior when critical measurements drop out
+
+## Deliverables
+Return:
+- measurement dictionary (name, unit, validity range, source)
+- validation and fallback rules
+- file/test changes
+- open instrumentation risks for commissioning
+
+Decision interview triggers:
+- changed units or semantics for released measurements
+- new fallback logic that may mask real field failures
+- altered quality thresholds affecting control decisions