From rtl-agent-team
Executes parallel Phase 5 RTL verification across multiple modules using Claude Code native teams. Manages 9 dependency-aware categories, task graphs, and graduation gates.
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>
Phase 4 completion required:
rtl/**/*.sv files must existreviews/phase-4-rtl/lint-report.md must exist (lint PASS)If prerequisites are missing: WARNING — recommend running /rtl-agent-team:rtl-p4-implement.
Proceed with available artifacts — orchestrator will adapt scope.
# Step 1: Team creation (main session = leader)
TeamCreate(team_name="p5-verify", description="Phase 5 verification: 9-category parallel verification")
# Step 2: Write team-config.json for hook consumption
Write(".rat/state/team-config.json", json.dumps({
"team_mode": true,
"team_name": "p5-verify",
"leader_session_id": "<current_session_id>",
"coordinator_name": "coordinator",
"worker_count": 4,
"phase": "p5",
"created_at": "<ISO_TIMESTAMP>"
}))
# Step 3: Prepare directories
Bash("mkdir -p docs/phase-5-verify reviews/phase-5-verify sim/coverage formal lint/cdc .rat/scratch/phase-5")
# Step 4: No initial tasks from skill — coordinator creates per-module V1-V9 after discovering modules
# Step 5: Spawn coordinator as teammate (orchestrator)
Agent(team_name="p5-verify", subagent_type="rtl-agent-team:p5-verify-team-orchestrator",
name="coordinator", description="P5 verification coordination",
prompt="You are the Phase 5 verification coordinator in team 'p5-verify'. "
"Manage the 9-category task graph using TaskCreate/TaskList/TaskUpdate. "
"Direct workers via SendMessage. "
"Create per-module V1-V9 task graph after discovering modules. "
"Signal leader when final compliance review complete. User input: $ARGUMENTS")
# Step 6: Spawn workers as teammates (4 general-purpose)
Agent(team_name="p5-verify", subagent_type="rtl-agent-team:func-verifier",
name="verify-0", description="P5 functional verification",
prompt="You are a Phase 5 verification worker in team 'p5-verify'. "
"Coordinator: 'coordinator' (send results via SendMessage). "
"Phase artifacts: reviews/phase-5-verify/, sim/. "
"Specialty: lint (V1), functional regression (V5), coverage (V6). "
"For specialist work, spawn: Task(subagent_type='rtl-agent-team:<specialist>', prompt='...'). "
"Examples: lint-checker for V1, testbench-dev for TB, eda-runner for sim, coverage-analyst for V6. "
"Follow Team Worker Protocol in agents/lib/team-worker-preamble.md.")
Agent(team_name="p5-verify", subagent_type="rtl-agent-team:sva-extractor",
name="verify-1", description="P5 formal and CDC verification",
prompt="You are a Phase 5 verification worker in team 'p5-verify'. "
"Coordinator: 'coordinator' (send results via SendMessage). "
"Phase artifacts: formal/, lint/cdc/, reviews/phase-5-verify/. "
"Specialty: SVA/formal (V2), CDC (V3), protocol (V4). "
"For specialist work, spawn: Task(subagent_type='rtl-agent-team:<specialist>', prompt='...'). "
"Examples: cdc-checker for V3, protocol-checker for V4, constraint-writer for SDC. "
"Follow Team Worker Protocol in agents/lib/team-worker-preamble.md.")
Agent(team_name="p5-verify", subagent_type="rtl-agent-team:eda-runner",
name="analysis-worker", description="P5 performance and synthesis",
prompt="You are a Phase 5 analysis worker in team 'p5-verify'. "
"Coordinator: 'coordinator' (send results via SendMessage). "
"Phase artifacts: reviews/phase-5-verify/. "
"Specialty: performance (V7), synthesis estimation (V8). "
"For specialist work, spawn: Task(subagent_type='rtl-agent-team:<specialist>', prompt='...'). "
"Examples: perf-verifier for V7, synthesis-reporter for V8. "
"Follow Team Worker Protocol in agents/lib/team-worker-preamble.md.")
Agent(team_name="p5-verify", subagent_type="rtl-agent-team:rtl-critic",
name="review-worker", description="P5 code review and compliance",
prompt="You are a Phase 5 review worker in team 'p5-verify'. "
"Coordinator: 'coordinator' (send results via SendMessage). "
"Phase artifacts: reviews/phase-5-verify/. "
"Specialty: code review (V9), requirement traceability, final compliance. "
"For specialist work, spawn: Task(subagent_type='rtl-agent-team:<specialist>', prompt='...'). "
"Examples: requirement-tracer for REQ mapping, rtl-critic for review. "
"Follow Team Worker Protocol in agents/lib/team-worker-preamble.md.")
# Step 7: Leader monitoring loop — poll until all tasks complete
while True:
tasks = TaskList()
all_done = all(t.status == "completed" for t in tasks)
if all_done:
break
# Continue polling
# Step 8: Cleanup
TeamDelete()
Bash("rm -f .rat/state/team-config.json")
Bash("rm -rf .rat/scratch/phase-5/")
acceptance_criteria are met.
When structured acceptance_criteria (with ac_id) exist from P3, verification
tracks compliance at the individual criterion level (ac_id granularity).
When acceptance_criteria is in string-array format (P1/P2), verification
operates at REQ level.acceptance_criteria