Yichen-ZJU/claude-for-research
Autonomous research for Claude Code: literature, ideas, unattended experiments, writing and review. 115 skills, two loops.
ARCHETYPE
Skill Collector
Ten-plus skills loaded on demand. A procedural-knowledge library.
Copy this rig
# review before running: this installs third-party code $ npx degit Yichen-ZJU/claude-for-research/agents ./rig-claude-for-research/agents $ npx degit Yichen-ZJU/claude-for-research/skills ./rig-claude-for-research/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 Yichen-ZJU/claude-for-research/skills/a-evolve .claude/skills/a-evolve $ npx degit Yichen-ZJU/claude-for-research/skills/academic-plotting .claude/skills/academic-plotting $ npx degit Yichen-ZJU/claude-for-research/skills/academic-research-suite .claude/skills/academic-research-suite $ npx degit Yichen-ZJU/claude-for-research/skills/accelerate .claude/skills/accelerate $ npx degit Yichen-ZJU/claude-for-research/skills/ai-use-disclosure .claude/skills/ai-use-disclosure $ npx degit Yichen-ZJU/claude-for-research/skills/alpha-research .claude/skills/alpha-research $ npx degit Yichen-ZJU/claude-for-research/skills/alphafold2 .claude/skills/alphafold2 $ npx degit Yichen-ZJU/claude-for-research/skills/anti-defensive-writing-en .claude/skills/anti-defensive-writing-en $ npx degit Yichen-ZJU/claude-for-research/skills/anti-defensive-writing .claude/skills/anti-defensive-writing $ npx degit Yichen-ZJU/claude-for-research/skills/autoresearch .claude/skills/autoresearch $ npx degit Yichen-ZJU/claude-for-research/skills/awq .claude/skills/awq $ npx degit Yichen-ZJU/claude-for-research/skills/benchmark-paper-template .claude/skills/benchmark-paper-template $ npx degit Yichen-ZJU/claude-for-research/skills/bitsandbytes .claude/skills/bitsandbytes $ npx degit Yichen-ZJU/claude-for-research/skills/blip-2 .claude/skills/blip-2 $ npx degit Yichen-ZJU/claude-for-research/skills/boltz .claude/skills/boltz $ npx degit Yichen-ZJU/claude-for-research/skills/borzoi .claude/skills/borzoi $ npx degit Yichen-ZJU/claude-for-research/skills/brainstorming-research-ideas .claude/skills/brainstorming-research-ideas $ npx degit Yichen-ZJU/claude-for-research/skills/chai1 .claude/skills/chai1 $ npx degit Yichen-ZJU/claude-for-research/skills/clip .claude/skills/clip $ npx degit Yichen-ZJU/claude-for-research/skills/compute-env-setup .claude/skills/compute-env-setup $ npx degit Yichen-ZJU/claude-for-research/skills/creative-thinking-for-research .claude/skills/creative-thinking-for-research $ npx degit Yichen-ZJU/claude-for-research/skills/customize .claude/skills/customize $ npx degit Yichen-ZJU/claude-for-research/skills/deep-research .claude/skills/deep-research $ npx degit Yichen-ZJU/claude-for-research/skills/deepspeed .claude/skills/deepspeed $ npx degit Yichen-ZJU/claude-for-research/skills/diffdock .claude/skills/diffdock $ npx degit Yichen-ZJU/claude-for-research/skills/docker .claude/skills/docker $ npx degit Yichen-ZJU/claude-for-research/skills/eli5 .claude/skills/eli5 $ npx degit Yichen-ZJU/claude-for-research/skills/esmfold2 .claude/skills/esmfold2 $ npx degit Yichen-ZJU/claude-for-research/skills/evo2 .claude/skills/evo2 $ npx degit Yichen-ZJU/claude-for-research/skills/experiment-forge .claude/skills/experiment-forge $ npx degit Yichen-ZJU/claude-for-research/skills/experiment-queue .claude/skills/experiment-queue $ npx degit Yichen-ZJU/claude-for-research/skills/experiment-watchdog .claude/skills/experiment-watchdog $ npx degit Yichen-ZJU/claude-for-research/skills/fair-esm2 .claude/skills/fair-esm2 $ npx degit Yichen-ZJU/claude-for-research/skills/figure-composer .claude/skills/figure-composer $ npx degit Yichen-ZJU/claude-for-research/skills/figure-style .claude/skills/figure-style $ npx degit Yichen-ZJU/claude-for-research/skills/flash-attention .claude/skills/flash-attention $ npx degit Yichen-ZJU/claude-for-research/skills/gguf .claude/skills/gguf $ npx degit Yichen-ZJU/claude-for-research/skills/gptq .claude/skills/gptq $ npx degit Yichen-ZJU/claude-for-research/skills/grpo-rl-training .claude/skills/grpo-rl-training $ npx degit Yichen-ZJU/claude-for-research/skills/hqq .claude/skills/hqq
$ curl -fsSL --create-dirs -o .claude/agents/researcher.md https://raw.githubusercontent.com/Yichen-ZJU/claude-for-research/main/agents/researcher.md $ curl -fsSL --create-dirs -o .claude/agents/reviewer.md https://raw.githubusercontent.com/Yichen-ZJU/claude-for-research/main/agents/reviewer.md $ curl -fsSL --create-dirs -o .claude/agents/verifier.md https://raw.githubusercontent.com/Yichen-ZJU/claude-for-research/main/agents/verifier.md $ curl -fsSL --create-dirs -o .claude/agents/writer.md https://raw.githubusercontent.com/Yichen-ZJU/claude-for-research/main/agents/writer.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 (114)
evolving-ai-agentsacademic-plottingacademic-research-suitehuggingface-accelerateai-use-disclosurealpha-researchalphafold2anti-defensive-writing-enanti-defensive-writingautoresearchawq-quantizationbenchmark-paper-templatequantizing-models-bitsandbytesblip-2-vision-languageboltzborzoibrainstorming-research-ideaschai1clipcompute-env-setupcreative-thinking-for-researchcustomizedeep-researchdeepspeeddiffdockdockereli5esmfold2evo2experiment-forgeexperiment-queueexperiment-watchdogfair-esm2figure-composerfigure-styleoptimizing-attention-flashgguf-quantizationgptqgrpo-rl-traininghqq-quantizationidea-evaluatorindication-dossierintro-drafterjobsknowledge-distillationligandmpnnliterature-reviewimplementing-llms-litgptllama-factoryllavaevaluating-llms-harnesslong-contextmanaged-model-endpointstraining-llms-megatronml-paper-writingml-training-recipemlflowmodal-computemodel-pruningmoe-trainingmonitor-experimentnanogptnemo-evaluator-sdkopenfold3openrlhf-trainingpaper-code-auditpaper-narrativepaper-productionpaper-writingpdf-explorepeft-fine-tuningpre-submission-reviewerpresenting-conference-talkspreviewproduct-self-knowledgeproteinmpnnpytorch-fsdp2remote-compute-modalremote-compute-sshreplicationresearch-orchestratorresearch-paper-writingresearch-reviewrun-experimentrunpod-computesaelensscgptscvi-toolssegment-anythingself-awarenesssession-logsession-searchsimposkill-creatorsolublempnnsource-comparisonstable-diffusionstyle-calibrationsudis-arr-rebuttal-with-mesudis-neurips-rebuttal-with-mesudis-rebuttal-with-mesummarizeswanlabsystematic-reviewsystems-paper-writingtech-paper-templatetensorboardtorchtitantransformer-lensunslothusing-model-endpointwatchweights-and-biaseswhisper
Subagents (4)
| researcher | Gather primary evidence across papers, web sources, repos, docs, and local artifacts. Use for evidence-gathering subtasks in research workflows (deep research, literature review, replication planning) |
| reviewer | Run tough but constructive internal research critique of a research artifact. Use as the adversarial verification/review pass after a cited draft exists. Flags FATAL/MAJOR/MINOR issues and writes a st |
| verifier | Post-process a draft to add inline citations and verify every source URL. Use after a draft exists in research workflows, before the review pass. Writes the complete cited document to a file. |
| writer | Turn research notes into clear, structured briefs and drafts. Use for the drafting/synthesis stage of research workflows when delegation is preferred over writing in the main session. Does NOT add cit |
Similar rigs
aipoch/open-science
The open-source AI research workbench for scientific research and agent workflows. Local-first, model-agnostic desktop app with extensible skills, MCP tools and connectors, Python/R execution and traceable artifacts for reproducible researc
Skill Collector 5.2k tok ·
cinicu/opencode-dotfiles
OpenCode dotfiles — 19 AI agent skills + 8 MCP servers + persistent memory (memory-keeper). Cross-platform install, screenshot OCR, custom CLI toolchain. Research methodology, systematic reviews, critical analysis, skill-creator.
Skill Collector 1.1k tok ·
wanshuiyin/Auto-claude-code-research-in-sleep
ARIS ⚔️ (Auto-Research-In-Sleep) — Lightweight Markdown-only skills for autonomous ML research: cross-model review loops, idea discovery, and experiment automation. No framework, no lock-in — works with Claude Code, Codex, OpenClaw, or any
Skill Collector 6.0k tok ·
juzser/blacksmith
Autonomous agent factory — loop runner, worktree engine, and a local dashboard over the event log.
Orchestrator 2.6k tok ·