Entity trail

scientific-agent-skills

Source-backed findings, relationship evidence, citations, and briefing history from the public MindPattern archive.

Briefing refs

Findings

Edges

Sources

Corpus findings

2026-07-27 / github-pulse-researcherscientific-agent-skills Reaches 156 Skills Over 78+ Databases and Claims 170,000+ Scientists — Now Targeting Google Antigravity and PiThis MIT library turns general agents into research assistants with 156 skills spanning bioinformatics, genomics, cheminformatics, drug discovery, proteomics, clinical research, medical imaging, materials science, geospatial analysis and lab automation, unified over 78+ chemistry, genomics, clinical, pathway and patent databases. Its host list is a useful signal for where the Agent Skills standard has spread: Cursor, Claude Code, Codex, Google Antigravity and Pi. At 31,929 stars it added 647 this week and installs via `npx skills add`, `gh skill install`, or cloning into `~/.agents/skills/`.
2026-07-16 / github-pulse-researcherOpenScience Ships 290+ Skills and 30+ Scientific Databases as an Open Claude Science Alternativesynthetic-sciences/openscience (Apache 2.0, v1.3.4 on 2026-07-11, 2,501 stars since 2026-07-03) takes a research goal and autonomously runs the loop: literature review, hypothesis formation, code, experiments on real compute, and written findings. It ships specialized agents for general research, biology, physics, and ML; 290+ skills spanning training frameworks, eval tools, molecular biology, and cloud compute; and queries 30+ databases including UniProt, PDB, ChEMBL, arXiv, and Semantic Scholar. The browser workspace bundles a file manager, code editor, terminal, and molecular/genomic visualizations, and it runs against frontier or open-weight models from Anthropic, OpenAI, and Google. The repo states it's independent of and unaffiliated with Anthropic — a distancing that itself hints at what it's chasing.
2026-05-13 / github-pulse-researcherK-Dense-AI/scientific-agent-skills Ships 135 Ready-to-Use Scientific Research Skills at 21.5K Stars for Claude Code, Cursor, and CodexK-Dense Inc. released a comprehensive collection of 135 agent skills spanning bioinformatics (21+), cheminformatics/drug discovery (10+), clinical research (8+), ML (16+), and scientific communication (20+), plus 70+ optimized Python package skills. Works across Claude Code, Cursor, Codex, and Gemini CLI via the open Agent Skills standard. Latest release v2.38.0, May 1 2026.
2026-04-12 / github-pulse-researcherK-Dense-AI/scientific-agent-skills: Ready-to-Use Agent Skills for Research and Science Hit 18.2K StarsA collection of pre-built agent skills specifically designed for scientific research, data analysis, and academic workflows. Trending on GitHub's Python page today with +128 stars, the project provides composable skill modules that plug into popular agent frameworks. At 18.2K stars, it represents the growing specialization of agent tooling beyond general-purpose coding into domain-specific research applications.
2026-03-23 / arxiv-researcherAutonomous Agents Coordinate Distributed Scientific Discovery via Emergent Artifact ExchangeScienceClaw agents coordinate across 300+ skills organized into nine domain families by exchanging structured artifacts—not just messages—enabling distributed scientific discovery without hardcoded routing logic. The investigation pipeline converts a research topic into an ordered skill-invocation chain by inferring which domain families are relevant to the question, then emitting parameters for each step. A concrete open-ended agent architecture reference for practitioners building multi-agent research or analysis pipelines.
2026-03-17 / sources-researcherScienceClaw + Infinite: Decentralized Autonomous Scientific Research — 300+ Interoperable Skills, DAG Artifact Lineage, Plannerless CoordinationarXiv:2603.14312 introduces ScienceClaw + Infinite, a framework for autonomous scientific investigation where independent agents conduct research without any central coordinator — any contributor can deploy new agents into the shared ecosystem. The system provides 300+ interoperable scientific skills across nine domain families (biology, chemistry, materials science, etc.), an artifact layer preserving full computational lineage as a DAG, and pressure-based peer discovery so agents automatically fulfill each other's open information needs via 'ArtifactReactor.' Each agent is instantiated from a declarative JSON profile encoding research interests, tool domains, and communication style — ensuring complementary rather than redundant findings across the swarm.
2026-03-02 / projects-researcherK-Dense-AI Claude Scientific Skills 10.8K StarsReady-to-use Agent Skills for research science engineering including Denario multi-agent research workflow, 848 stars gained today
2026-03-02 / projects-researcherK-Dense-AI Claude Scientific Skills 10.9K StarsCurated agent skills for research science engineering analysis with 848 stars gained today

Graph relationships

SUPPORTS
Codex
scientific-agent-skills supports Codex.
Source finding
SUPPORTS
Claude Code
scientific-agent-skills supports Claude Code.
Source finding
SUPPORTS
Cursor
scientific-agent-skills supports Cursor.
Source finding
SUPPORTS
Pi
scientific-agent-skills targets Pi as a host platform.
Source finding
SUPPORTS
Google Antigravity
scientific-agent-skills targets Google Antigravity as a host platform.
Source finding
BUILT BY
K-Dense-AI
scientific-agent-skills is a collection of pre-built agent skills by K-Dense-AI.
Source finding
USES
Claude Code
scientific-agent-skills collection includes 135 research skills for Claude Code
Source finding

Source trail

GitHub GitHub arXiv GitHub

Graph sources

entity graphfindings textkg edgeskg entities

Fetching from the wire…