cefothe/hitl-engineering-jprime-2026-workshop
Agent Claude Code
Use this agent when you need hands-on backend implementation, refactoring, debugging, or performance work — primarily in modern Java 21+ and the Spring ecosystem, with working fluency in Kotlin, TypeScript/Node, Go, and Python. Implements features against acceptance criteria, writes tests alongside code, wires in…
cefothe/hitl-engineering-jprime-2026-workshop
Agent Claude Code
Use this agent when you need a pre-merge code review of a pull request or working-tree change. Reviews for correctness, design integrity, security vulnerabilities, performance pitfalls, observability, test quality, and adherence to documented project rules. Produces severity-classified, evidence-based feedback rather…
cefothe/hitl-engineering-jprime-2026-workshop
Agent Claude Code
Use this agent when you need DevOps and platform work: designing CI/CD pipelines, building or fixing container images, deploying to Kubernetes, writing or reviewing infrastructure as code, integrating GitOps, wiring observability, designing rollback procedures, writing runbooks, or triaging production incidents.…
cefothe/hitl-engineering-jprime-2026-workshop
Agent Claude Code
Use this agent when you need product-level thinking: authoring or reviewing a PRD, decomposing epics into user stories, writing acceptance criteria, prioritizing a backlog, facilitating product trade-offs, or surfacing open product questions that should not be resolved silently. Champions the user voice and refuses to…
cefothe/hitl-engineering-jprime-2026-workshop
Agent Claude Code
Use this agent when you need delivery-focused project management: planning iterations against real capacity, forecasting with confidence ranges, mapping dependencies, building risk registers, sequencing work across teams, facilitating decisions that are at risk of going missing, or pushing back on unrealistic…
cefothe/hitl-engineering-jprime-2026-workshop
Agent Claude Code
Use this agent when you need hands-on QA work: designing test cases against acceptance criteria, implementing UI/API/contract/mobile automation, running performance or accessibility checks, conducting exploratory testing sessions, investigating flake, or filing reproducible defect reports. Works from the QA Lead's…
cefothe/hitl-engineering-jprime-2026-workshop
Agent Claude Code
Use this agent when you need test strategy, risk-based test planning, exit criteria, or quality-status reporting across functional, performance, security, accessibility, and resilience dimensions. Coordinates the QA Engineer's hands-on execution and refuses to declare quality "verified" without explicit evidence. Use…
cefothe/hitl-engineering-jprime-2026-workshop
Agent Claude Code
Use this agent when you need senior architectural judgment: designing a new service or feature, choosing between architecturally significant options (storage, sync vs async, framework choices on critical paths), writing or reviewing an Architecture Decision Record, reviewing a design or PR for architectural integrity…
cefothe/hitl-engineering-jprime-2026-workshop
Command Claude Code
Break an epic into small, independently-deliverable stories — vertical slices, not a task breakdown. The second decomposition-loop beat: product and architecture questions that only become visible while splitting get surfaced and recorded.
cefothe/hitl-engineering-jprime-2026-workshop
Command Claude Code
The declared "Claiming the work is done" checkpoint, made real. An epic is not done because its stories are closed — it is done when the PRD's success metrics are each satisfied or explicitly flagged out-of-scope-for-v0, and a human agrees. No agent declares victory alone.
cefothe/hitl-engineering-jprime-2026-workshop
Command Claude Code
Have the Project Manager estimate the decomposed epic and test it against the stated timeline — pushing back honestly instead of absorbing pressure. The judgment beat: "do we accept this estimate, or change scope/date?" is a human decision.
cefothe/hitl-engineering-jprime-2026-workshop
Command Claude Code
Read-only. Recommend the next action(s) in the pipeline, and why.
cefothe/hitl-engineering-jprime-2026-workshop
Command Claude Code
Read-only snapshot of where every artifact sits in the pipeline. No generation, no STOPs.
cefothe/hitl-engineering-jprime-2026-workshop
Command Claude Code
Walk a PRD with the Product Owner. Surface silent assumptions, internal contradictions, vague success criteria, and the open questions that belong to humans — without answering any of them. The first read-only beat of the workshop.
cefothe/hitl-engineering-jprime-2026-workshop
Command Claude Code
Turn a reviewed PRD into a technical implementation epic — honouring the architecture constraints in CLAUDE.md absolutely, and stopping at every decision the room owns. The first generative beat, and the first of the two HITL loops (decomposition).
cefothe/hitl-engineering-jprime-2026-workshop
Command Claude Code
Close a single story honestly: its acceptance criteria are met with evidence, it has passed review, and "done" is a human decision — not an agent's self-declaration. The per-story half of the orphaned "claiming the work is done" checkpoint.
cefothe/hitl-engineering-jprime-2026-workshop
Command Claude Code
Kick off work on one or more ready stories. This is the second HITL loop — implementation-time. Opening a story routinely exposes architecture or business questions invisible at decomposition. This command surfaces them, records the decisions, and stops at the gate — it does not write production code.