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-biological-process-engineering/SKILL.md

@@ -0,0 +1,54 @@
+---
+name: evolv-biological-process-engineering
+description: Engineer biological wastewater process behavior for EVOLV nodes. Use when implementing or reviewing reactor/settler biology, ASM-style kinetics, oxygen demand, nitrification/denitrification, sludge behavior, calibration assumptions, and biologically plausible constraints.
+---
+
+# EVOLV Biological Process Engineering
+
+## Mission
+Keep EVOLV biological process models physically plausible, calibratable, and operationally useful.
+
+## Harness Execution Contract
+- Ground changes in current biology/state variables and connected control topics.
+- Define invariants before edits:
+  - biological mass-balance intent is preserved
+  - model assumptions remain explicit and traceable
+  - degraded behavior remains availability-first and bounded
+- Validate with deterministic tests and representative operating scenarios.
+
+## Scope
+- `nodes/reactor/`
+- `nodes/settler/`
+- `nodes/pumpingStation/` (where biology interacts with flow/retention assumptions)
+- Related reaction modules and utilities under `nodes/*/src/`
+
+## Workflow
+1. Identify biological state variables, units, and expected ranges.
+2. Map kinetic pathways (growth, decay, transfer, conversion) and rate constraints.
+3. Verify oxygen/temperature dependencies and fallback behavior.
+4. Check integration stability for configured time-step and resolution choices.
+5. Confirm outputs remain interpretable for control and dashboard consumers.
+
+## Standards
+- Keep state vectors explicit and index mappings documented.
+- Avoid silent clipping/coercion without test coverage and rationale.
+- Preserve topic/payload compatibility unless migration is defined.
+- Record calibration assumptions and required field data.
+
+## Test Expectations
+Cover:
+- kinetic branch behavior under representative and boundary conditions
+- non-negativity and boundedness safeguards
+- temperature and oxygen transfer sensitivity
+- time-step/resolution edge behavior and stability warnings
+
+## Deliverables
+Return:
+- biological assumptions and constraints used
+- changed files/tests and evidence
+- calibration notes and unresolved biological uncertainties
+
+Decision interview triggers:
+- altered biology assumptions that can change plant behavior
+- parameter/default changes with startup or compliance impact
+- compatibility breaks in biological outputs or topic contracts