nw-agent-builder
Use when creating new AI agents, validating agent specifications, optimizing requested existing SKILL.md assets, optimizing command definitions, or ensuring compliance with Claude Code best practices. Creates focused, research-validated agents (200-400 lines) with Skills for domain knowledge.
Wave: Other Model: inherit Max turns: 40 Tools: Read, Write, Edit, Glob, Grep, Task, SendMessage
Commands
Preloaded skills
- nw-ab-agent-template — KNOWLEDGE — the canonical agent-spec template (frontmatter + body skeleton). Reference loaded by the create/migrate procedures; no sequence.
- nw-ab-anti-patterns — KNOWLEDGE — agent/skill/command anti-pattern catalog with fixes. Reference scanned by validate-spec; no sequence.
- nw-ab-critique-dimensions — Review dimensions for validating agent quality - template compliance, safety, testing, and priority validation
- nw-ab-examples — KNOWLEDGE — canonical worked examples for agent creation, migration, and command optimization. Reference for the relevant procedures; no sequence.
- nw-ab-house-style — KNOWLEDGE — caveman-native authoring house style + by-construction guarantees (Reasoning Mandate injection, measured-gain compression). Reference for create/migrate; no sequence.
- nw-ab-validation-checklist — KNOWLEDGE (data) — the agent-spec validation checklist. The item definitions the validate-spec procedure RUNS against. No sequence of its own.
- nw-agent-creation-workflow — Detailed 5-phase workflow for creating agents - from requirements analysis through validation and iterative refinement
- nw-agent-testing — 5-layer testing approach for agent validation including adversarial testing, security validation, and prompt injection resistance
- nw-cross-cutting-invariants — Cross-cutting normative invariants — routing core for the software/model boundary doctrine, gate/construction principles and on-demand knowledge lenses. Cite clause ids; never re-declare.
- nw-design-patterns — 7 agentic design patterns with decision tree for choosing the right pattern for each agent type
- nw-typesafe-system-one — Use Jev System One for every supported, authorized semantic judgment when available: relevance selection, handoff preflight, evidence mapping, finding clustering, and confidence-gated escalation.