Research & Data Agents tools for Claude Code

Wiki-centric full-lifecycle research platform built on Claude Code, realizing Karpathy's LLM-Wiki vision. 20+ skills cover the full loop: ingest → ideate → novelty check → experiment design / run / eval → paper writing. Research state lives in a structured knowledge wiki with an interactive graph

by Pedro H. C. Sant'Anna - A ready-to-fork Claude Code template for academics using LaTeX/Beamer + R. Multi-agent review, quality gates, adversarial QA, and replication protocols. Great use of orchestration patterns and great documentation

ShinkaEvolve: Towards Open-Ended and Sample-Efficient Program Evolution

Open-source research workspace with sequential idea-to-paper pipelines and integrated autoresearch tool packs

CORAL: Autonomous multi-agent evolution for open-ended discovery (arXiv:2604.01658). Long-running agents with shared persistent memory, asynchronous execution, and heartbeat-based interventions; SOTA on 10 math/algorithmic/systems tasks

by Wenyu Chiou - A catalog of 15 Claude Code skills mapped to 8 research-workflow stages (literature → gap analysis → design → drafting → reviewer response), where each stage emits an explicit YAML/Markdown deliverable the next stage consumes. Thoughtfully designed around anti-hallucination — schemas force "gap" status on unsupported claims, and downstream skills refuse malformed, overconfident ha

Fully autonomous AI scientist built on Claude Code, with explicit AutoResearch lineage, multi-agent research iteration, GPU experiment execution, and a self-evolving outer loop

Self-improving software factory for AI/ML objectives. State an objective and it runs a campaign: candidate solutions designed, implemented by coding agents (Claude Code, Codex), measured against the objective, and the closest refined until it is met. Each finished campaign leaves lessons carrying the evidence that earned them, and repositories and papers feed the same knowledge hub, so the next ca

by muend - Skills for geospatial analysis that require coordinate-system, spatial-validation and comparability checks before a result is reported, and narrow or refuse a claim the data cannot support. Ships a paired evaluation against a skills-disabled control arm

Claude Code skill for public SEC EDGAR + market data: /analyze, /score, /compare. MIT

Multi-agent QA toolkit with 10 specialized agents — orchestrator, environment-manager, functional-reviewer, test-scenario-designer, browser-validator, automation-writer, manual-validator, bug-reporter, release-analyzer, smart-test-selector. Stack-agnostic, output-chained, live validation via Chrome MCP

Zero-config AI agent CLI with 20+ tools, persistent memory, Ollama auto-detection, and free tier

Claude Code skill that runs an autoresearch-style, human-gated deep-learning loop across six roles, versioned experiments, and evidence-checked completion