- 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>
2.4 KiB
2.4 KiB
name, description
| name | description |
|---|---|
| evolv-process-systems-control | Design and review system-level control behavior across EVOLV process nodes. Use when coordinating multi-node interactions, mode/state transitions, parent-child registration flows, control loops, and complex process sequencing spanning reactors, valves, pumps, settlers, and machine groups. |
EVOLV Process Systems Control
Mission
Preserve stable system-wide behavior across interacting Node-RED nodes and process assets.
Harness Execution Contract
- Build a topic and ownership map from the current repo before changing behavior.
- Define invariants before editing:
- no unplanned break in released topic contracts
- explicit safe defaults and transition guards
- deterministic output sequencing assumptions
- Return concrete evidence (tests/trace examples) for sequence and fail-safe claims.
Scope
- Cross-node interactions via
msg.topic - Parent-child registration contracts (
registerChildand related topics) - Mode management and sequencing in node wrappers/domain classes
Message/Port Convention Baseline
Many EVOLV nodes use this output convention:
- output 0: process message
- output 1: database/influx message
- output 2: parent/registration/control plumbing
Preserve topic stability once released (registerChild, setMode, setScaling, etc). If a topic contract changes, define a migration path.
Control Workflow
- Map control boundaries and authority (who commands whom).
- List topic contracts and payload schemas.
- Verify state/mode transition logic for race/conflict conditions.
- Define safe startup, shutdown, and failover behavior.
- Confirm tick timing and output ordering assumptions.
Design Rules
- Keep topic names stable once released.
- Use explicit transition guards and default-safe modes.
- Avoid hidden cross-coupling between unrelated assets.
- Make control intent observable in outputs/status.
Test Expectations
Add tests for:
- normal sequence transitions
- out-of-order messages
- duplicate child registration and dedupe behavior
- fail-safe behavior under missing dependencies
Deliverables
Return:
- system interaction map (topics + ownership)
- transition table and safety guards
- changed files/tests
- unresolved control hazards with mitigation suggestions
Decision interview triggers:
- any topic rename/removal or payload schema break
- authority changes across parent/child nodes
- startup/shutdown sequencing changes with operational impact