Role Max — Testing & QA agent for Claude Code
A fondue role (Architect, Architecture Reviewer, Coder, Code Reviewer, Tests Auditor, Conventions Auditor, Handover Writer in a spec; Hunter, Reviewer in a hunt) or an Advisor, at max effort.
How to install Role Max
Installs to ~/.claude/agents/jsobolewski1-ensemble-role-max.md
mkdir -p ~/.claude/agents && curl -fsSL https://raw.githubusercontent.com/jsobolewski1/ensemble/HEAD/agents/role-max.md -o ~/.claude/agents/jsobolewski1-ensemble-role-max.md Restart Claude Code, or start a new session, for it to be picked up.
What Role Max does
name: role-max description: A fondue role (Architect, Architecture Reviewer, Coder, Code Reviewer, Tests Auditor, Conventions Auditor, Handover Writer in a spec; Hunter, Reviewer in a hunt) or an Advisor, at max effort. Spawned only by a fondue Arbiter; never delegate other work to it. effort: max experimental: cacheTtl: 1h
You hold one role in a fondue spec or hunt, or you advise its User. If your first message names a skill, a role and a folder, load that skill and follow it. If it s
Alternatives in Testing & QA
- Facet Coder — Generates ONE facet Cypress spec (properties propertiesFx styles stylesFx events csa inspector variants) for a 41k ★
- Track Plan — Track ID: {{TRACK_ID}} Spec: spec.md Estimated Effort: {{EFFORT_ESTIMATE}} 31.9k ★
- Fg QA Advisor — Finance Guru Quality Assurance Advisor (Dr. Jennifer Wu) 317 ★
Full documentation available on GitHub
View Source RepositoryRelated Agents
Role High
A fondue role (Architect, Architecture Reviewer, Coder, Code Reviewer, Tests Auditor, Conventions Auditor, Han
Role Low
A fondue role (Architect, Architecture Reviewer, Coder, Code Reviewer, Tests Auditor, Conventions Auditor, Han
Role Medium
A fondue role (Architect, Architecture Reviewer, Coder, Code Reviewer, Handover Writer in a spec; Hunter, Revi
Fleet Architect
Fleet Architect of the fleetsmith fleet for Meta agent-fleet builder: one fleet.yaml spec compiles into coordi
Tester Glm
GLM 5.3 Flash/max test and verification agent. Use when the GLM lane is selected; OpenAI uses qa/verify and Cl
Factory Coder
Agent Factory role — makes the Tester's failing test GREEN with the minimal, contract-first production change.
Related Skills
Fable Skills
Five skills Fable wrote before it left: the security sweep, the setup, the build planner, the honest advisor,
Hunt Android
Spawn parallel Android hunter agents against an acquired and extracted app
Paper Ref
Hunt candidate references in the current project's references library (and optionally the web) for a given top