Wh Designer — Testing & QA agent for Claude Code
Design-phase agent.
How to install Wh Designer
Installs to ~/.claude/agents/muhibm1-workhorse-snapshot-wh-designer.md
mkdir -p ~/.claude/agents && curl -fsSL https://raw.githubusercontent.com/muhibm1/workhorse-snapshot/HEAD/agents/wh-designer.md -o ~/.claude/agents/muhibm1-workhorse-snapshot-wh-designer.md Restart Claude Code, or start a new session, for it to be picked up.
What Wh Designer does
name: wh-designer description: Design-phase agent. In one session turns " -> " into the risk tier, spec.md with ADRs, evals.md within the tier's case limit, plan.md in file-disjoint waves where every eval case has exactly one task, and brief.md, the Design document the human approves at G2. Never stops to ask; every question becomes a decision row with a recommendation. Use when a change starts, or to revise once after a high audit finding or a G2 rejection. tools: Read, Gl
Alternatives in Testing & QA
- Eval Author — Authoring role for the D365FO agent eval loop catalog 136 ★
- Eval Auditor — Audits an evaluation setup (benchmark, A/B test, or model comparison) for methodology errors that would invali 51 ★
- QA Strategist — Use this agent for read-only test-coverage gap analysis between waves 49 ★
Full documentation available on GitHub
View Source RepositoryRelated Agents
AI Eval Designer
Use this agent to design a risk-tiered evaluation set for an AI feature. Trigger when the user says "design ev
Spgr Agent Backend Developer
Implements server-side features strictly within the confirmed API spec, ERD, ADRs, and tech stack, test-first,
Stx Dev Base
Universal Dev agent prelude for /stx-feature waves. Every tier-specialized Dev (db / service / api / ui) loads
Skill Verifier
Fresh-session surrogate verifier for improve skill rewrites. Use when /improve reaches Step 4.5 and needs isol
Coverage Strategist
Decides what the suite should test and what it should stop testing, within the twenty case cap. Use before add
Opus Agent
Tier opus worker. Work carrying a decision — rule or spec prose that states a rule, an exported signature or m
Related Skills
Claude Evals
Production eval framework for Claude Agent SDK — implements Anthropic's published eval patterns with native SD
Architect Prove
Run the prove stage. Writes at least ten eval cases including a refusal and an injection case, runs the harnes
AI Risk Review
Prompt: AI Governance — Risk Tier & Eval Coverage