From rtl-agent-team
Enforces full RTL bug fix cycle: analyze → fix → lint → testbench create/update → functional verification. Prevents lint-only completions for .sv files.
npx claudepluginhub babyworm/rtl-agent-team --plugin rtl-agent-teamThis skill uses the workspace's default tool permissions.
<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.
Core principle: lint is only a syntax check, not evidence of functional correctness.
This skill integrates with the PostToolUse:Edit hook tracking system. Modified .sv files are automatically tracked, and session termination is blocked without functional verification.
<Use_When>
<Do_Not_Use_When>
<Why_This_Exists> Born from a previous session where 9 RTL bugs were fixed with only lint runs, skipping TB/simulation entirely. Passing lint is merely "compilation success" — simulation is required to prove the fix is functionally correct.
Anti-pattern example:
All policies, checklists, escalation rules, and parallel fix decision logic are defined in the rtl-p4s-bugfix-policy skill (loaded via the orchestrator's skills: field).
Task(subagent_type="rtl-agent-team:p4s-bugfix-orchestrator", prompt="Execute RTL bug fix cycle. Bug description: $ARGUMENTS")
Do not perform any work directly. The orchestrator agent manages the full analyze → fix → lint → TB → simulation cycle.
.rat/scratch/phase-4/bugfix-decision-{N}.md — decision record documenting root cause, fix rationale, and verification evidence