From rtl-agent-team
Executes Phase 1-3 RTL design pipeline from spec to uArch using parallel native teams within each phase and inter-phase quality gates. Use for maximum intra-phase parallelism.
npx claudepluginhub babyworm/rtl-agent-team --plugin rtl-agent-teamThis skill is limited to using the following tools:
<Purpose>
Generates design tokens/docs from CSS/Tailwind/styled-components codebases, audits visual consistency across 10 dimensions, detects AI slop in UI.
Records polished WebM UI demo videos of web apps using Playwright with cursor overlay, natural pacing, and three-phase scripting. Activates for demo, walkthrough, screen recording, or tutorial requests.
Delivers idiomatic Kotlin patterns for null safety, immutability, sealed classes, coroutines, Flows, extensions, DSL builders, and Gradle DSL. Use when writing, reviewing, refactoring, or designing Kotlin code.
<Use_When>
<Do_Not_Use_When>
No phase prerequisites (starts from Phase 1).
Specification documents should be available in specs/ directory.
# Legacy migration: rename pre-0.6.10 state file ONLY if new file does not exist
legacy = Read(".rat/state/rtl-spec-to-uarch-state.json") # may not exist
if legacy:
new_exists = Read(".rat/state/rat-p1p3-spec-uarch-state.json") # check new file
if not new_exists:
Bash("mv .rat/state/rtl-spec-to-uarch-state.json .rat/state/rat-p1p3-spec-uarch-state.json")
# Initialize or resume state
state = Read(".rat/state/rat-p1p3-spec-uarch-state.json") # may not exist
if state and state.current_phase > 1:
# Resume: skip completed phases
pass
else:
# Fresh start
Write(".rat/state/rat-p1p3-spec-uarch-state.json",
{ "schema_version": "3.0", "current_phase": 1, "pipeline_scope": "phase-1-to-3",
"execution_mode": "team",
"phases": {
"1": { "status": "pending" },
"2": { "status": "pending" },
"3": { "status": "pending" }
}
})
# ── Phase 1: Research (team) ──────────────────────────────────
if phases["1"]["status"] != "completed":
Skill(skill="rtl-agent-team:rtl-p1-research-team", args="$ARGUMENTS")
# Phase 1→2 artifact gate (iron-requirements is the settled SSOT;
# open-requirements is optional — absent if Phase 1 had no deferred research)
Glob("docs/phase-1-research/iron-requirements.json")
Glob("docs/phase-1-research/io_definition.json")
Glob("docs/phase-1-research/timing_constraints.json")
Glob("docs/phase-1-research/domain-analysis.md")
# All four required artifacts must exist. If missing: FAIL.
# Update state
# phases["1"]["status"] = "completed"
# ── Phase 2: Architecture + RefC (team) ───────────────────────
if phases["2"]["status"] != "completed":
Skill(skill="rtl-agent-team:rtl-p2-arch-team", args="Phase 1 artifacts complete")
# Phase 2→3 artifact gate
Glob("docs/phase-2-architecture/architecture.md")
Glob("refc/**/*.c")
# Must exist. If missing: FAIL.
# Update state
# phases["2"]["status"] = "completed"
# ── Phase 3: uArch + BFM (team) ──────────────────────────────
if phases["3"]["status"] != "completed":
Skill(skill="rtl-agent-team:rtl-p3-uarch-team", args="Phase 2 artifacts complete")
# Phase 3 artifact gate
Glob("docs/phase-3-uarch/*.md")
Glob("bfm/")
# Must exist. If missing: FAIL.
# Update state
# phases["3"]["status"] = "completed"
# ── Completion ────────────────────────────────────────────────
# Report: Phase 1-3 artifacts, reviews, ADR count
# Suggest: "Run /rtl-agent-team:rat-p4p5-impl-verify to begin RTL implementation"
# Do NOT proceed to Phase 4
Each phase's team skill handles its own TeamCreate/TeamDelete lifecycle. This skill sequences phases and enforces inter-phase quality gates.