Runs automated design space exploration loops for hardware and EDA workflows. You give it a program (gem5, yosys, verilator), parameters to sweep (cache sizes, pipeline widths, optimization flags), and an objective metric, and it iteratively runs experiments, parses results, and picks the next configuration to try. Stops when it hits your target, times out, or sees no improvement for N iterations. Smart about inferring parameter ranges from source code if you don't specify them, like reading argparse definitions or Verilog parameters to figure out reasonable values. Built for the architecture and verification crowd who'd otherwise write bash loops and spreadsheets. Already has internal looping so don't wrap it in cron.

skills.sh installs
519
GitHub stars
18.9K
last updated
Aug 21, 2026

Install

$ npx -y skills add wanshuiyin/auto-claude-code-research-in-sleep --skill dse-loop --agent claude-code

Installs into .claude/skills of the current project.

Files

DSE Loop: Autonomous Design Space Exploration

πŸ” Do not wrap this skill in /loop / CronCreate. It already loops internally until its objective is met or it times out. Unlike the verdict-bearing review/audit skills, its stop gate is an objective machine-checkable metric (Type-A), so its self-termination is safe same-model β€” the reason not to wrap it is scheduler duplication, not the verdict fence. See shared-references/external-cadence.md.

Autonomously explore a design space: run β†’ analyze β†’ pick next parameters β†’ repeat, until the objective is met or timeout is reached. Designed for computer architecture and EDA problems.

Context: $ARGUMENTS

Safety Rules β€” READ FIRST

NEVER do any of the following:

  • sudo anything
  • rm -rf, rm -r, or any recursive deletion
  • rm any file you did not create in this session
  • Overwrite existing source files without reading them first
  • git push, git reset --hard, or any destructive git operation
  • Kill processes you did not start

If a step requires any of the above, STOP and report to the user.

Constants (override via $ARGUMENTS)

Sponsors

Details

Repository
github.com/wanshuiyin/auto-claude-code-research-in-sleep
Categories
Code Review & Quality, Data Science & ML, Design & UI/UX
First seen
Jun 3, 2026

More from wanshuiyin/auto-claude-code-research-in-sleep

All 41
1motion-team avatarvisual-formatsmotion-team/creative-strategy-skillsA reference library of 45+ visual ad formats for Meta and paid social, each with a definition, funnel stage guidance, and medium (video/static/both). Use this w516skills.sh installs
2longbridge avatarlongbridge-financial-checkuplongbridge/skillsSystematic 100-point financial health scorecard for listed companies via Longbridge β€” five dimensions: profitability (ROE / net margin / gross margin), growth (515skills.sh installs
3dotneet avatarprompt-improverdotneet/claude-code-marketplaceA skill for improving prompts by applying general LLM/agent best practices. When the user provides a prompt, this skill outputs an improved version, identifies 514skills.sh installs
4manutej avatargrpc-microservicesmanutej/luxor-claude-marketplaceComprehensive gRPC microservices skill covering protobuf schemas, service definitions, streaming patterns, interceptors, load balancing, and production gRPC arc510skills.sh installs
5aj-geddes avatartechnical-debt-assessmentaj-geddes/useful-ai-promptsAssess, quantify, and prioritize technical debt using code analysis, metrics, and impact analysis. Use when planning refactoring, evaluating codebases, or makin509skills.sh installs
6dalestudy avatarskill-creatordalestudy/skillsDaleStudy/skills μ €μž₯μ†Œμ˜ μŠ€ν‚¬ 생성, μˆ˜μ •, 검증 μ‹œ μ‚¬μš©. λ‹€μŒ μƒν™©μ—μ„œ ν™œμ„±ν™”: (1) μƒˆ μŠ€ν‚¬ 생성 μš”μ²­ μ‹œ, (2) skills/ 디렉토리 λ‚΄ SKILL.md 파일 μˆ˜μ • μ‹œ, (3) SKILL.md λ³€κ²½ κ²€ν†  λ˜λŠ” 리뷰 μ‹œ, (4) μŠ€ν‚¬ frontmatter λ˜λŠ” 509skills.sh installs