babyworm/rtl-agent-team/rtl-p3-uarch-team

Claude Code and agent skill by babyworm/rtl-agent-team

Trust tier 3
  • community
Source check pending

Checking that the SKILL.md file is still in the repository.

Phase 3 uArch with parallel team workers: dual-stream per-block uArch + BFM development. Use for 'uarch team', 'Phase 3 team', 'parallel uarch'.

Install

npx vskill@latest install babyworm/rtl-agent-team/rtl-p3-uarch-team

SKILL.md

<Purpose> Execute Phase 3 uArch design pipeline using Claude Code native team infrastructure. The skill (main session) handles team lifecycle: TeamCreate, coordinator + worker spawning, task monitoring, and cleanup. The coordinator teammate manages task graphs and directs workers via SendMessage. </Purpose>

<Use_When>

  • Phase 2 architecture is complete and uArch design is needed
  • User says "uarch team", "Phase 3 team", "parallel uarch"
  • Have multiple modules requiring parallel uarch design
  • Need maximum parallelism for dual-stream uArch + BFM development

</Use_When>

<Do_Not_Use_When>

  • Phase 2 architecture not complete (run p2-arch-design first)
  • Single module only (use rtl-p3-uarch-design for simpler flow)
  • Only need BFM (use bfm-develop)

</Do_Not_Use_When>

Prerequisites

Phase 2 completion required:

  • docs/phase-2-architecture/architecture.md must exist
  • refc/ directory with C reference model must exist

If prerequisites are missing: WARNING — recommend running /rtl-agent-team:p2-arch-design. Proceed with available artifacts — orchestrator will adapt scope.

Execution

python
# Step 1: Team creation (main session = leader)
TeamCreate(team_name="p3-uarch", description="Phase 3 uArch: dual-stream uArch + BFM development")
# Step 2: Write team-config.json for hook consumption
Write(".rat/state/team-config.json", json.dumps({
"team_mode": true,
"team_name": "p3-uarch",
"leader_session_id": "<current_session_id>",
"coordinator_name": "coordinator",
"worker_count": 3,
"phase": "p3",
"created_at": "<ISO_TIMESTAMP>"
}))
# Step 3: Prepare directories
Bash("mkdir -p docs/phase-3-uarch reviews/phase-3-uarch .rat/scratch/phase-3")
# Step 4: Create initial task graph (T1 and T2 are independent — no dependencies)
TaskCreate(subject="T1: uArch Design",
description="Per-block micro-architecture design from architecture spec")
TaskCreate(subject="T2: BFM Development",
description="Bus Functional Model development from architecture spec")
# Step 5: Spawn coordinator as teammate (orchestrator)
Agent(team_name="p3-uarch", subagent_type="rtl-agent-team:p3-uarch-team-orchestrator",
name="coordinator", description="P3 uArch coordination",
prompt="You are the Phase 3 uArch coordinator in team 'p3-uarch'. "
"Manage the task graph using TaskCreate/TaskList/TaskUpdate. "
"Direct workers via SendMessage. "
"Initial tasks T1 (uarch) and T2 (BFM) already created by leader. "
"Create dynamic tasks (T3 BFM gate, T4-T10 review rounds), then perform Step 3.7/3.8 "
"iron artifact production (REQ-U-* from resolved OPEN-2-*) + open resolution + "
"zero-opens + ambiguity gate, then Step 3.9 compliance. "
"Signal leader ONLY after the Phase 3 Gate (Step 4) passes — including the "
"Step 3.9 compliance check, ADR generation, and the mandatory Codex cross-review. "
"User input: $ARGUMENTS")
# Step 6: Spawn workers as teammates (3 general-purpose)
Agent(team_name="p3-uarch", subagent_type="rtl-agent-team:uarch-designer",
name="uarch-worker", description="P3 uArch design and review",
prompt="You are a Phase 3 uArch worker in team 'p3-uarch'. "
"Coordinator: 'coordinator' (send results via SendMessage). "
"Phase artifacts: docs/phase-3-uarch/, reviews/phase-3-uarch/. "
"Specialty: per-block uArch design, timing review, algorithm consistency. "
"For specialist work, spawn: Task(subagent_type='rtl-agent-team:<specialist>', prompt='...'). "
"Examples: timing-advisor for timing review, vcodec-architecture-expert for domain. "
"Follow the Team Worker Protocol section of your agent definition. "
"Scratch dir: .rat/scratch/phase-3/ (for write-restricted outputs).")
Agent(team_name="p3-uarch", subagent_type="rtl-agent-team:bfm-dev",
name="bfm-worker", description="P3 BFM development",
prompt="You are a Phase 3 BFM worker in team 'p3-uarch'. "
"Coordinator: 'coordinator' (send results via SendMessage). "
"Phase artifacts: bfm/, docs/phase-3-uarch/. "
"Specialty: BFM development, BFM correctness review, I/O logging. "
"For specialist work, spawn: Task(subagent_type='rtl-agent-team:ref-model-dev', prompt='...'). "
"Follow the Team Worker Protocol section of your agent definition.")
Agent(team_name="p3-uarch", subagent_type="rtl-agent-team:rtl-architect",
name="review-worker", description="P3 review lead",
prompt="You are a Phase 3 review worker in team 'p3-uarch'. "
"Coordinator: 'coordinator' (send results via SendMessage). "
"Phase artifacts: reviews/phase-3-uarch/. "
"Specialty: feature preservation review, aggregation, final consolidation. "
"For specialist work, spawn: Task(subagent_type='rtl-agent-team:<specialist>', prompt='...'). "
"Follow the Team Worker Protocol section of your agent definition. "
"Scratch dir: .rat/scratch/phase-3/ (for write-restricted outputs).")
# Step 7: Leader wait — teardown is gated on the COORDINATOR SIGNAL, not on TaskList.
# "All tasks completed" is necessary but NOT sufficient: the coordinator's post-gate
# work (compliance check, ADR generation, phase summary, Codex cross-review) runs via
# direct Task() calls that never appear in the shared task graph.
while True:
if coordinator_signaled_phase_complete:
# SendMessage from 'coordinator' — sent only after the phase gate
# AND all post-gate mandatory steps pass (see coordinator prompt above)
break
tasks = TaskList()
if all(t.status == "completed" for t in tasks):
# Task graph drained but no coordinator signal yet → the coordinator is
# still running gate/post-gate steps. KEEP WAITING — do NOT TeamDelete.
pass
# Continue polling
# Step 8: Cleanup
TeamDelete()
Bash("rm -f .rat/state/team-config.json")
Bash("rm -rf .rat/scratch/phase-3/")

Workflow Notes

  • Open Resolution: resolve all OPEN-2-* items from Phase 2 open-requirements.json
  • Zero-Opens Invariant: no unresolved OPEN-* items may pass to Phase 4
  • Exit gate includes open-resolved, zero-remaining-opens, and ambiguity-pass
  • Exit gate includes compliance-pass (compliance-checker vs P1+P2 iron, orchestrator Step 3.9)

Security

  • Security scanPassed, 95 of 100Pattern checks on the skill's files.
  • Model reviewPassed

Quality evaluation

  • Task evaluationNot runNo evaluation yet. Run one in Skill Studio.

Source

Popularity

7-day trend score
0

Not refreshed yet.

Works with

Any agent vskill installs to: 56 targets in all. These 39 have a page on this site.

  • Claude Code
  • Cursor
  • GitHub Copilot
  • Windsurf
  • Codex
  • Gemini CLI
  • Cline
  • Amp
  • Roo Code
  • Goose
  • Continue
  • Kilo Code
  • Junie
  • Trae
  • OpenHands
  • Qwen Code
  • AI Agent
  • Tabnine Chat
  • Sourcegraph Cody
  • Amazon Q
  • Firebase Genkit
  • Sweep
  • Devin
  • Phind
  • Mentat
  • Claude Desktop
  • Aider
  • GPT Pilot
  • Plandex
  • bolt.new
  • v0
  • Replit Agent
  • Anse
  • TypingMind
  • BetterChatGPT
  • LibreChat
  • Warp AI
  • Pieces
  • Fig AI
Embed badge
vskill VERIFIED badge
[![vskill VERIFIED](https://verified-skill.com/api/v1/skills/babyworm/rtl-agent-team/rtl-p3-uarch-team/badge)](https://verified-skill.com/skills/babyworm/rtl-agent-team/rtl-p3-uarch-team)

Something wrong with this skill or its page? Tell us.