How do I install the Rtl To Signoff skill?

Run npx skillmds@latest add together2329/rtl-to-signoff in your terminal (requires Node.js), paste this page's agent-chat prompt into Claude, Cursor, or any MCP-connected agent, or download the SKILL.md file and copy it into your agent's skills directory.

Is Rtl To Signoff safe to use?

This skill has not completed SkillMD's automated safety review yet. SkillMD never runs a skill's scripts for you; review the SKILL.md before installing.

Which AI agents work with Rtl To Signoff?

This skill is tagged as working with Claude Code, Claude.ai, OpenAI Codex. SKILL.md is an open format, so most agents that read a skills directory can load it too.

Is Rtl To Signoff free to use?

Yes. Installing skills from SkillMD is free, and the skill stays under its author's original license.

Who published Rtl To Signoff?

together2329 (@together2329) published this skill. Their other Agent Skills are listed on their SkillMD profile.

Rtl To Signoff by together2329 — Agent Skill

Name: Rtl To Signoff
Author: together2329

Rtl To Signoff

Execute or plan the real ATLAS RTL-to-signoff flow from RTL disk checks through TB, sim, lint, coverage, synthesis, STA, PnR, and post-route STA. Use when the user asks for RTL to SignOff, RTL signoff, or full hardware closure.

together2329 Updated Sep 17, 2026

File contents

RTL To Signoff

This is the Cursor entrypoint for a real RTL-to-signoff loop. It follows common_ai_agent workflow semantics:

DV/equivalence stages call WorkflowStageEngine.run_stage().
EDA stages call workflow command scripts such as auto_syn.sh and auto_sta.sh.
Stage ownership, evidence paths, and profiles come from STAGE_MANIFEST.json.
The generated summary is evidence aggregation, not a replacement validator.
Sim-debug evidence includes FL-vs-RTL classification plus sim/simulation_quality.json from scoreboard semantic checks.

Read First

SIGNOFF_RUNBOOK.md
workflow/COMMON_ENGINE_FLOW.md
doc/wiki/workflow-ownership-and-boundaries.md
doc/wiki/golden-todo-evidence.md

Plan The Flow

From common_ai_agent/, inspect the commands without running them:

python3 .cursor/skills/rtl-to-signoff/scripts/rtl_to_signoff.py <ip> --profile full --plan

If the IP lives outside common_ai_agent/, pass the IP parent:

python3 .cursor/skills/rtl-to-signoff/scripts/rtl_to_signoff.py <ip> --root .. --profile full --plan

Execute

Run DV/equivalence closure:

python3 .cursor/skills/rtl-to-signoff/scripts/rtl_to_signoff.py <ip> --root . --profile dv --execute

Run EDA closure after DV is green:

python3 .cursor/skills/rtl-to-signoff/scripts/rtl_to_signoff.py <ip> --root . --profile eda --execute

Run the full path:

python3 .cursor/skills/rtl-to-signoff/scripts/rtl_to_signoff.py <ip> --root . --profile full --execute

Run from SSOT generation when the IP does not yet have an approved SSOT:

python3 .cursor/skills/rtl-to-signoff/scripts/rtl_to_signoff.py <ip> --root . --profile ssot-to-signoff --provider fake --execute

Rules

Fix failures through the owner shown by the manifest or summary; do not edit evidence to pass.
Treat pre-existing green JSON as stale after RTL/TB/SSOT changes.
After TB, sim, RTL, SSOT, or equivalence-goal changes, refresh sim-debug quality evidence before final closure.
EDA stages may require Linux, PDK paths, Yosys, OpenSTA, OpenROAD, Verilator, and tool env from .env.example.
Final claims must cite fresh logs/stage_engine/*.json, EDA command output, or verify/cursor_rtl_to_signoff_summary.json.

---name: rtl-to-signoffdescription: Execute or plan the real ATLAS RTL-to-signoff flow from RTL disk checks through TB, sim, lint, coverage, synthesis, STA, PnR, and post-route STA. Use when the user asks for RTL to SignOff, RTL signoff, or full hardware closure.---# RTL To SignoffThis is the Cursor entrypoint for a real RTL-to-signoff loop. It follows common_ai_agent workflow semantics:- DV/equivalence stages call `WorkflowStageEngine.run_stage()`.- EDA stages call workflow command scripts such as `auto_syn.sh` and `auto_sta.sh`.- Stage ownership, evidence paths, and profiles come from `STAGE_MANIFEST.json`.- The generated summary is evidence aggregation, not a replacement validator.- Sim-debug evidence includes FL-vs-RTL classification plus `sim/simulation_quality.json` from scoreboard semantic checks.## Read First- `SIGNOFF_RUNBOOK.md`- `workflow/COMMON_ENGINE_FLOW.md`- `doc/wiki/workflow-ownership-and-boundaries.md`- `doc/wiki/golden-todo-evidence.md`## Plan The FlowFrom `common_ai_agent/`, inspect the commands without running them:```bashpython3 .cursor/skills/rtl-to-signoff/scripts/rtl_to_signoff.py <ip> --profile full --plan```If the IP lives outside `common_ai_agent/`, pass the IP parent:```bashpython3 .cursor/skills/rtl-to-signoff/scripts/rtl_to_signoff.py <ip> --root .. --profile full --plan```## ExecuteRun DV/equivalence closure:```bashpython3 .cursor/skills/rtl-to-signoff/scripts/rtl_to_signoff.py <ip> --root . --profile dv --execute```Run EDA closure after DV is green:```bashpython3 .cursor/skills/rtl-to-signoff/scripts/rtl_to_signoff.py <ip> --root . --profile eda --execute```Run the full path:```bashpython3 .cursor/skills/rtl-to-signoff/scripts/rtl_to_signoff.py <ip> --root . --profile full --execute```Run from SSOT generation when the IP does not yet have an approved SSOT:```bashpython3 .cursor/skills/rtl-to-signoff/scripts/rtl_to_signoff.py <ip> --root . --profile ssot-to-signoff --provider fake --execute```## Rules- Fix failures through the owner shown by the manifest or summary; do not edit evidence to pass.- Treat pre-existing green JSON as stale after RTL/TB/SSOT changes.- After TB, sim, RTL, SSOT, or equivalence-goal changes, refresh sim-debug quality evidence before final closure.- EDA stages may require Linux, PDK paths, Yosys, OpenSTA, OpenROAD, Verilator, and tool env from `.env.example`.- Final claims must cite fresh `logs/stage_engine/*.json`, EDA command output, or `verify/cursor_rtl_to_signoff_summary.json`.

Frequently asked questions