This repository also configures its own agents. See what wize-development-kit tells them →
Command Claude CodeOpenCode
Goal. Surface evidence that hardens the brief and trigger map. Frame the questions Pepper actually needs answered, do focused passes (market / competitors / analytics / interviews), and write the synthesis in a way Maria Hill can quote in the PRD.
Command Claude CodeOpenCode
Goal. Capture what worked, what didn't, what surprised, and one concrete change for next sprint. A retro that ends in a vibe is a retro that didn't ship a change.
Command Claude CodeOpenCode
Red-team review of an artifact. Used by Hawkeye on stories/architecture and by Peggy on prose.
Command Claude CodeOpenCode
Splits large markdown (PRD, architecture, etc.) into addressable shards under {doc-name}.shards/.
Command Claude CodeOpenCode
Goal. Distill any intent input into a canonical, preservation-validated machine contract for downstream Wize work.
Command Claude CodeOpenCode
4-implementation: Sprint Planning.
Command Claude CodeOpenCode
4-implementation: Sprint Status.
Command Claude CodeOpenCode
Goal. For one story: declare the test split, the fixtures, the mocks, the environments, and the assertion targets. This is Hawkeye's contract Shuri implements against. No story enters dev without one.
Command Claude CodeOpenCode
Goal. Final per-story decision. PASS / CONCERNS / FAIL / WAIVED with documented rationale, score, and policy mode. This is the line: a story doesn't advance past it without a recorded gate.
Command Claude CodeOpenCode
Goal. Verify that the epic as a whole meets Fury's NFR principles: performance, security, reliability, maintainability, accessibility, cost. Story-level gates miss NFRs that emerge only at integration; this gate catches them.
Command Claude CodeOpenCode
Goal. A structured review of one story before its gate decision. Distinct from Shuri's wize-code-review (which audits code health). This one audits AC fulfillment, test discipline, and risk-spot coverage.
Command Claude CodeOpenCode
Goal. Build the probability × impact matrix that prioritizes the rest of TEA's work. Areas in HIGH get deep test design; areas in LOW get smoke. Without a risk profile, every story is tested the same — wasteful, and unsafe.
Command Claude CodeOpenCode
Goal. Map every Acceptance Criterion to one or more concrete tests in the repo. Reports honest coverage: covered, partial, missing. A story with missing rows cannot pass gate.
Command Claude CodeOpenCode
2-to-3-boundary: Tech Vision.
Command Claude CodeOpenCode
1-analysis: Technical Research.
Command Claude CodeOpenCode
Goal. Map user psychology to business goals. For each user action the product wants to drive, name the trigger (the need/anxiety/desire that precedes it), the friction that blocks it today, the business outcome unlocked, and the signal that confirms it happened.
Command Claude CodeOpenCode
Goal. Turn scenarios into page-by-page specifications Tony can architect against and Shuri can implement. One spec per screen. Every spec answers: purpose, primary action, layout, content, interactions, states (loading/empty/error/success/disabled), accessibility, and metrics.
Command Claude CodeOpenCode
Goal. Produce a scenario map that anchors all UX work. Eight focused questions; one short answer each. Scenarios are pieces of behavior, not personas — they're how this user, in this state, tries to accomplish that JTBD.
Command Claude CodeOpenCode
Goal. Decide whether the PRD is ready to leave Plan and enter Solutioning. Catch ambiguity before architecture is wasted on it. Cheap step; saves multiples of itself downstream.
At most 3 mods per repository are shown here, and a mod shipped inside a plugin is left to that plugin's page — the rest are on their repository pages: