Generate deterministic publication-quality architecture, workflow, and pipeline diagrams from structured JSON (FigureSpec) into editable SVG. Use when user says "架构图", "workflow 图", "pipeline 图", "确定性矢量图", "figure spec", "draw architecture", or needs precise, editable, publication-ready vector diagrams. Preferred over AI illustration for formal architecture/workflow figures.
idea-discovery
idea-discovery is an open-source workflow skill for Claude Code and compatible agents, published by wanshuiyin. Its author describes it as: “Workflow 1: Full idea discovery pipeline to go from a broad research direction to validated, pilot-tested ideas. Use when user says "找idea全流程", "idea discovery pipeline", "从零开始找方向", or wants the complete idea explorat…”. The project has 15k stars on GitHub and is available under the MIT license. Add it to your setup with `git clone https://github.com/wanshuiyin/Auto-claude-code-research-in-sleep ~/.claude/skills/idea-discovery`.
What idea-discovery does
Orchestrate a complete idea discovery workflow for: **$ARGUMENTS**
Installation
Add idea-discovery to your agent with:
git clone https://github.com/wanshuiyin/Auto-claude-code-research-in-sleep ~/.claude/skills/idea-discoveryAlways review a skill's source before installing it. This command comes from the skill's public repository; the linked repo is the source of truth for exact setup steps.
What's inside
The SKILL.md for idea-discovery is organised into these sections:
- Overview
- Constants
- Per-stage evidence gate (RESUMABLE = true)
- Pipeline
- Phase 0: Load Research Brief (if available)
- Phase 0.5: Reference Paper Summary (when REFPAPER is set)
- Phase 1: Literature Survey
- Phase 2: Idea Generation + Filtering + Pilots
- Phase 3: Deep Novelty Verification
- Phase 4: External Critical Review
- Phase 4.5: Method Refinement + Experiment Planning
- Phase 5: Final Report
When to use it
Reach for idea-discovery when you want workflow help from your agent without writing the same instructions every session. Load the skill and the agent picks it up automatically for relevant tasks.
Strengths
- Clear MIT license — safe to read and adapt
- Ships in wanshuiyin/Auto-claude-code-research-in-sleep, an established project with 14,858 GitHub stars
- Actively maintained (recent commits)
Topics
Frequently asked questions
- What does idea-discovery do?
- Workflow 1: Full idea discovery pipeline to go from a broad research direction to validated, pilot-tested ideas. Use when user says "找idea全流程", "idea discovery pipeline", "从零开始找方向", or wants the complete idea exploration workflow.
- How do I install idea-discovery?
- Run git clone https://github.com/wanshuiyin/Auto-claude-code-research-in-sleep ~/.claude/skills/idea-discovery in your agent, then reload your skills. Review the source at https://github.com/wanshuiyin/Auto-claude-code-research-in-sleep before installing.
- Is idea-discovery free to use?
- Yes. idea-discovery is free and open source under the MIT license, so you can read, run, and adapt it within that license's terms.
- Where does idea-discovery come from?
- idea-discovery ships inside wanshuiyin/Auto-claude-code-research-in-sleep, a repository that contains 41 catalogued skills in total. The repository's 14,858 GitHub stars apply to that whole collection, not to this skill on its own.
Related skills
More Workflow →Workflow 1 adaptation for robotics and embodied AI. Orchestrates robotics-aware literature survey, idea generation, novelty check, and critical review to go from a broad robotics direction to benchmark-grounded, simulation-first ideas. Use when user says "robotics idea discovery", "机器人找idea", "embodied AI idea", "机器人方向探索", "sim2real 选题", or wants ideas for manipulation, locomotion, navigation, drones, humanoids, or general robot learning.
When you can't trust yourself with your code base, trust Arbiter.
Provides PyTorch-native distributed LLM pretraining using torchtitan with 4D parallelism (FSDP2, TP, PP, CP). Use when pretraining Llama 3.1, DeepSeek V3, or custom models at scale from 8 to 512+ GPUs with Float8, torch.compile, and distributed checkpointing.
Parameter-efficient fine-tuning for LLMs using LoRA, QLoRA, and 25+ methods. Use when fine-tuning large models (7B-70B) with limited GPU memory, when you need to train <1% of parameters with minimal accuracy loss, or for multi-adapter serving. HuggingFace's official library integrated with transformers ecosystem.