One-click install
npx skills add https://github.com/raja21068/AutoResearch --skill dse-loop-raja21068
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: dse-loop
Source: https://github.com/raja21068/AutoResearch/tree/main/skills/aris/dse-loop
Command: npx skills add https://github.com/raja21068/AutoResearch --skill dse-loop-raja21068

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

This Skill removes the manual trial-and-error of parameter tuning by running an execution-grounded design space exploration loop until your optimization goal is met or time runs out.

Core Features & Use Cases

  • Run → Analyze → Tune → Iterate: Executes a program, parses objective metrics from outputs, selects the next parameter configuration, and repeats while tracking the best result.
  • Automatic range inference: If you provide parameter names without ranges, it reads the project/source code to infer valid defaults and sets conservative, domain-reasonable sweep ranges (with logging for review).
  • Safe logging and recovery: Writes each evaluated configuration to dse_results/dse_log.csv, maintains state in dse_results/DSE_STATE.json, and supports resuming after context/window compaction.
  • Typical Use Cases: microarchitecture DSE (gem5), synthesis/EDA tuning (yosys/DC, openroad/innovus), simulation parameter sweeps (verilator), compiler flag optimization (gcc/llvm), and formal verification tuning (abc/sby).

Quick Start

Use the dse-loop skill to run your benchmark while tuning the specified parameters, optimizing the chosen objective, and stopping when the target condition is satisfied or the timeout is reached.

Frequently Asked Questions about dse-loop

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I automate parameter tuning for computer architecture simulations?

Automate iterative parameter tuning by running a design space exploration loop that executes your program, extracts objective metrics, and selects new parameter configurations based on prior results until optimization goals are met or timeout is reached.

What is the best way to run a parameter sweep with constraints for EDA workflows?

Run an execution-grounded design space exploration loop to perform parameter sweeps with constraints for EDA workflows, automatically logging each evaluated configuration and maintaining state to support recovery and reproducibility.

Do I need to specify parameter ranges manually for iterative optimization?

You do not need to specify parameter ranges manually; the exploration loop can read your project source code to infer valid defaults and set conservative, domain-reasonable sweep ranges automatically when ranges are missing.

How does design space exploration handle context window compaction during long runs?

Design space exploration handles context window compaction by writing each evaluated configuration to a log file and maintaining state in a JSON file, allowing the iterative optimization loop to resume automatically after interruptions.

Can I use this for compiler flag optimization and formal verification tuning?

You can use this for compiler flag optimization and formal verification tuning across tools like gcc, llvm, abc, and sby, applying iterative parameter sweeps to optimize your chosen objective metric.