zingspark/create-harness-vibe-coding

zingspark/create-harness-vibe-coding

Agent Claude Code

WF-MAX Manager for W1 architecture wave. Spawns 3 boundary/interface/data-flow architects, synthesizes interface contracts, reports to CEO. Read-only + Agent spawn; no Edit/Write.

zingspark/create-harness-vibe-coding

Agent Claude Code

Use to review layer boundaries, ports, data flow, state machines, dependency direction, and architecture impact before implementation.

zingspark/create-harness-vibe-coding

Agent Claude Code

Use for scoped read-only source exploration, file discovery, symbol tracing, and simple codebase summaries before planning or implementation.

zingspark/create-harness-vibe-coding

Agent Claude Code

Use when context exceeds 85% of the window, or during WF closeout to extract durable knowledge before compression. Read-only analysis except for writing compression suggestions to PROGRESS.md#Heartbeat.

zingspark/create-harness-vibe-coding

Agent Claude Code

Use to isolate a failing command, reproduce the smallest failing path, and propose the narrowest fix.

zingspark/create-harness-vibe-coding

Agent Claude Code

Use to verify official documentation, APIs, SDK behavior, config options, version changes, limits, error semantics, and examples before implementation.

zingspark/create-harness-vibe-coding

Agent Claude Code

WF-MAX Manager for W0 exploration wave. Spawns 5-10 read-only researchers/explorers, synthesizes findings, reports to CEO. Read-only + Agent spawn; no Edit/Write.

zingspark/create-harness-vibe-coding

Agent Claude Code

WF-MAX Manager for W2 implementation wave. Spawns 5-7 implementers (one fileclaim each), merges results, reports to CEO. Agent spawn + synthesis only; does NOT write code directly.

zingspark/create-harness-vibe-coding

Agent Claude Code

Use to implement the smallest change inside a declared write set after tests or manual checks are defined.

zingspark/create-harness-vibe-coding

Agent Claude Code

Use when a repeated failure, user correction, or WF recovery loop triggers memory writing; also use during WF closeout for consolidation. Writes to Harness/memory/, Harness/MEMORY.md, and cross-project global memory.

zingspark/create-harness-vibe-coding

Agent Claude Code

Use to split a goal into tasks, dependencies, write sets, verification steps, and a Parallel Dispatch table before multi-step work.

zingspark/create-harness-vibe-coding

Agent Claude Code

Use after verification and cross-review to synthesize findings, detect unresolved risk, and decide whether work may enter final acceptance.

zingspark/create-harness-vibe-coding

Agent Claude Code

Use for product, market, competitor, open-source, dependency, pricing, policy, or ecosystem research before PRD and architecture decisions.

zingspark/create-harness-vibe-coding

Agent Claude Code

Optional native WF-REVIEW manager. Selects a bounded reviewer fan-out by risk, deduplicates evidence, assigns severity, and reports to the controller. Read-only + native Agent spawn; no Edit/Write or external CLI.

zingspark/create-harness-vibe-coding

Agent Claude Code

Use for read-only review of diffs, tests, architecture boundaries, regressions, and missing verification before closeout.

zingspark/create-harness-vibe-coding

Agent Claude Code

Use to maintain Harness task state, heartbeat, dispatch ledger, evidence pointers, and compact PLAN/PROGRESS updates. Small-fast chore agent. Never edits source code.

zingspark/create-harness-vibe-coding

Agent Claude Code

Test-Driven Development specialist enforcing AC-linked write-tests-first methodology. Use PROACTIVELY when writing new features, fixing bugs, refactoring code, or adding browser-visible behavior.

zingspark/create-harness-vibe-coding

Agent Claude Code

Use to define or write AC-linked failing tests and manual verification steps before implementation.

zingspark/create-harness-vibe-coding

Agent Claude Code

Use to run verification commands, inspect results, and record evidence. Final acceptance still waits for cross-review PASS and reflector PASS.