junjiezhou1122/ideascientist-claude-code
A native Claude Code translation of IdeaScientist for automatic, evidence-grounded research ideation.
ARCHETYPE
Orchestrator
A bench of specialised subagents. The main agent mostly delegates.
Copy this rig
# review before running: this installs third-party code $ npx degit junjiezhou1122/ideascientist-claude-code/agents ./rig-ideascientist-claude-code/agents $ npx degit junjiezhou1122/ideascientist-claude-code/skills ./rig-ideascientist-claude-code/skills
MCP servers are added to Claude Code at local scope; env vars are shown as YOUR_… placeholders — we never store values. Files are fetched with degit into a separate folder so you can review before merging.
$ npx degit junjiezhou1122/ideascientist-claude-code/skills/check .claude/skills/check $ npx degit junjiezhou1122/ideascientist-claude-code/skills/evaluate .claude/skills/evaluate $ npx degit junjiezhou1122/ideascientist-claude-code/skills/ideate .claude/skills/ideate $ npx degit junjiezhou1122/ideascientist-claude-code/skills/resume .claude/skills/resume
$ curl -fsSL --create-dirs -o .claude/agents/blind-baseline.md https://raw.githubusercontent.com/junjiezhou1122/ideascientist-claude-code/main/agents/blind-baseline.md $ curl -fsSL --create-dirs -o .claude/agents/gap-finder.md https://raw.githubusercontent.com/junjiezhou1122/ideascientist-claude-code/main/agents/gap-finder.md $ curl -fsSL --create-dirs -o .claude/agents/innovator.md https://raw.githubusercontent.com/junjiezhou1122/ideascientist-claude-code/main/agents/innovator.md $ curl -fsSL --create-dirs -o .claude/agents/orchestrator.md https://raw.githubusercontent.com/junjiezhou1122/ideascientist-claude-code/main/agents/orchestrator.md $ curl -fsSL --create-dirs -o .claude/agents/paper-reader.md https://raw.githubusercontent.com/junjiezhou1122/ideascientist-claude-code/main/agents/paper-reader.md $ curl -fsSL --create-dirs -o .claude/agents/report-writer.md https://raw.githubusercontent.com/junjiezhou1122/ideascientist-claude-code/main/agents/report-writer.md $ curl -fsSL --create-dirs -o .claude/agents/reviewer.md https://raw.githubusercontent.com/junjiezhou1122/ideascientist-claude-code/main/agents/reviewer.md
This rig commits no guardrails. Here is the community baseline instead — the deny/ask rules most often found across all 7,204 rigs:
{
"permissions": {
"deny": [
"Read(./.env)",
"Read(**/.env)",
"Read(~/.ssh/**)",
"Bash(rm -rf *)",
"Read(**/*.pem)",
"Bash(rm -rf /)",
"Bash(git push --force:*)",
"Bash(sudo *)",
"Read(.env)",
"Bash(rm -rf /*)",
"Read(./.env.*)",
"Read(~/.aws/**)",
"Bash(git push --force*)",
"Bash(rm -rf:*)",
"Read(**/*.key)",
"Read(**/.env.*)",
"Bash(sudo:*)",
"Bash(git reset --hard*)",
"Bash(git reset --hard:*)",
"Read(.env.*)"
],
"ask": [
"Bash(git push:*)",
"Bash(git push *)",
"Bash(git commit:*)",
"Bash(rm *)",
"Bash(rm:*)",
"Bash(git rebase *)",
"Bash(wget *)",
"Bash(npm publish:*)",
"Bash(git commit *)",
"Bash(gh pr merge *)"
]
}
} Skills (4)
Subagents (7)
| blind-baseline model: inherit | Freezes mechanisms derived from only the problem and near-miss digests before a proposal exists. |
| gap-finder model: inherit | Finds zero to two grounded methodological gaps on exactly one assigned challenge axis. |
| innovator model: inherit | Produces one sourced intuition for one grounded gap, or a new ID for a reviewed revision. |
| orchestrator model: inherit | Coordinates a bounded native research run from stage artifacts. Use for standalone ideation or resume orchestration. |
| paper-reader model: inherit | Reads one registered exact paper version and appends results-masked goal answers to its exclusive digest. |
| report-writer model: inherit | Develops the exact current surviving intuition into all fourteen proposal fields. |
| reviewer model: inherit | Actively disconfirms a candidate, or independently evaluates a complete native proposal. |
Similar rigs
fjerbi/claude-code-arsenal
A tested Claude Code operating system for efficient, reliable, and focused software engineering. Includes global instructions, workflows, rules, and tools designed to minimize wasted tokens, prevent hallucinations and unnecessary loops, enf
YOLO Cowboy 4.0k tok ·
IntuitivePhella/AI-PhellOS
Portable AI development workflow template and CLI for spec-driven, test-aware software delivery with Codex, Claude Code, and other coding agents.
Orchestrator 3.1k tok ·
TechHU-GS/cc-discipline
Guardrails for Claude Code. Shell hooks that actually block bad behavior — edit loops, skipped verification, destructive git — not just markdown that asks nicely.
Automator 7.6k tok ·
Clarittyai/agentic-app-seed
Self-hostable template for building AI workflow automations and agentic apps: FastAPI, React and Postgres, wired for Claude Code. Docker-ready.
Pragmatist 11.3k tok ·