upstream/GoekeLab-awesome-genomic-skills/catalogue/README.md

@@ -2,9 +2,9 @@
 title: "Awesome Genomic Skills [![Awesome](https://awesome.re/badge.svg)](https://awesome.re)"
 task: ""
 lineage_type: import
-upstream_source: https://github.com/GoekeLab/awesome-genomic-skills/blob/8cf42e1d/README.md
-upstream_sha: 8cf42e1d
-imported_at: 2026-08-08
+upstream_source: https://github.com/GoekeLab/awesome-genomic-skills/blob/281f0b83/README.md
+upstream_sha: 281f0b83
+imported_at: 2026-10-26
 prompt_class: catalogue
 upstream_changes: accepted
 author: upstream
@@ -23,15 +23,12 @@ A curated list of skills and MCP servers for working with AI coding agents (Clau
 
 ## Contents
 
-- [Awesome Genomic Skills ](#awesome-genomic-skills-)
-    - [What is a Skill and what is an MCP?](#what-is-a-skill-and-what-is-an-mcp)
-  - [Contents](#contents)
 - [Bioinformatics and Genomics Agent Skills](#bioinformatics-and-genomics-agent-skills)
 - [MCP Servers for Life Sciences](#mcp-servers-for-life-sciences)
+- [AI Scientist and Research Agent Systems](#ai-scientist-and-research-agent-systems)
 - [Benchmarks](#benchmarks)
 - [General AI Coding Agent Skill Collections](#general-ai-coding-agent-skill-collections)
 - [Other Notable Awesome Lists!](#other-notable-awesome-lists)
-  - [Contributing](#contributing)
 
 ## Bioinformatics and Genomics Agent Skills
 
@@ -93,8 +90,17 @@ Existing registries and lists of MCP servers:
 - [awesome-mcp-servers](https://github.com/punkpeye/awesome-mcp-servers#bio) - General-purpose MCP server list with a Biology, Medicine, and Bioinformatics subsection.
 
 MCP related tools: 
+- [Paper2Agent](https://github.com/jmiao24/Paper2Agent) - A multi-agent system that auto-generates an MCP server from a paper's codebase by executing its tutorials and extracting the workflows as validated, parameterised tools. [Nature paper](https://doi.org/10.1038/s41586-026-11044-y)
 - [BioinfoMCP](https://github.com/florensiawidjaja/BioinfoMCP) - Not strictly an MCP, a converter that auto-generates MCP servers from existing tool documentation, plus a benchmark of the converted tools. Preprint available [here](https://arxiv.org/abs/2510.02139)
 
+## AI Scientist and Research Agent Systems
+
+Standalone multi-agent applications that run end-to-end research workflows (e.g. hypothesis generation), rather than components you plug into your own agent.
+
+- [Co-Scientist](https://github.com/Kaimen-Inc/Co-Scientist)
+  - **Description:** Independent implementation of Google DeepMind's AI co-scientist; takes a natural-language research goal and returns a tournament-ranked overview of novel hypotheses via six agents (Generation, Reflection, Elo-tournament Ranking, Evolution, Proximity, Meta-review) under a Supervisor. Grounds hypotheses with PubMed, arXiv, Europe PMC, and web search (with URL provenance logging), and uses a pinned copy of DeepMind's [science-skills](https://github.com/google-deepmind/science-skills). Runs on an LLM API key (OpenRouter, Anthropic, OpenAI, Gemini) or a Claude Code / Codex subscription; includes a cross-model benchmarking harness. General-science rather than genomics-specific. Apache-2.0.
+  - **Developers:** Kaimen Inc, co-founded by Kuan-Lin Huang ([Huang Lab | Precision Omics](https://labs.icahn.mssm.edu/kuanhuanglab/), Icahn School of Medicine at Mount Sinai); not affiliated with Google or the AI co-scientist paper's authors.
+
 ## Benchmarks
 
 Benchmarks, evaluations, and other helpful resources at the intersection of AI and genomics/bioinformatics.