From rtl-agent-team
Executes Phase 4-to-5 RTL implementation and verification pipeline from existing Phase 1-3 design documents. Produces RTL code, unit tests, and full verification with feedback loops. Trigger: 'implement and verify'.
How this skill is triggered — by the user, by Claude, or both
Slash command
/rtl-agent-team:rat-p4p5-impl-verify [resume or module-name][resume or module-name]This skill is limited to the following tools:
The summary Claude sees in its skill listing — used to decide when to auto-load this skill
<Purpose>
<Use_When>
<Do_Not_Use_When>
Phase 3 completion with review required:
docs/phase-3-uarch/ must contain at least one μArch specreviews/phase-3-uarch/uarch-review.md must existIf prerequisites are missing: WARNING — recommend completing Phase 1-3
(via /rtl-agent-team:rat-p1p3-spec-uarch). Proceed with available artifacts — orchestrator will adapt scope.
Task(subagent_type="rtl-agent-team:uarch-to-verify-orchestrator", prompt="Execute Phase 4→5 RTL implementation and verification pipeline. User input: $ARGUMENTS")
Do not perform any work directly. The orchestrator agent manages prerequisite verification, dual-stream Phase 4, sub-phase Phase 5, feedback loops, and state management.
npx claudepluginhub babyworm/rtl-agent-team --plugin rtl-agent-teamImplements SystemVerilog RTL modules from μArch specs through a 10-wave pipeline: write, lint, fix, review, bugfix, unit test, CDC, protocol, refactor, integration gate.
Orchestrates parallel SystemVerilog RTL design builds: decomposes into independent modules, spawns code-gen agents, lints, verifies components, and integrates top-level. Use /gf-build for RISC-V CPUs or similar.
Creates structured, bite-sized implementation plans from specs or requirements before writing code. Useful for breaking down multi-step tasks into testable steps with file structure and task boundaries.