Orchestrates the autonomous legacy-to-Quarkus M-process (M1–M5) — analyzes findings, sequences stories, specifies per story, dispatches rewrite/infer work with sensors, and ships through the factory gate. Use in harness sessions when executing any M-stage of…
Agentic quality-advance gate for Stage 080 Track B. Use after every T-NNN task commit (detailed analysis — not a one-liner), after every major milestone M (M1–M5), before M5 ship/push, before story-complete / next story, after abort, and on escalations.…
Author and improve rhoai3-coding-demo documentation: root README, stage READMEs, docs/OPERATIONS.md, docs/TROUBLESHOOTING.md, BACKLOG.md, and supporting docs. Use when creating or updating demo prose, adding architecture sections, introducing platform…
Safely gather OpenShift cluster state for the RHOAI demo. Use when troubleshooting deployment failures, validating stage completion, or gathering diagnostic information. Must be used in readonly mode — no mutations. Do NOT use for remediation (use…
Manage shared agent guidance for this project — AGENTS.md, shared rules, shared skills, hooks, and tool bridges. Use when the user asks to create a rule, update a skill, audit rules, review skills, add a hook, create a subagent, or asks about .agents/ or…
Scale models and GPU MachineSets up or down in the RHOAI demo environment. Use when the user wants to stop/start models, scale GPU nodes, save costs, manage cluster resources, reduce cloud spend overnight, or shut down GPU capacity. For resuming Stage 020/030…
Use when creating, planning, implementing, or reviewing a new rhoai3-coding-demo stage from ideation to validated GitOps implementation. Covers stage scope, dependencies, skill routing, README drafting, GitOps ownership decisions, Argo CD Application setup,…
Resume the Stage 020/030 GPUaaS and private model-serving path after GPU nodes were intentionally scaled to zero, the demo environment was shut down, or private models are pending after GPU capacity returns. Use when the user asks to start the demo after…