Traction scoreGitHub stars can be faked, so popularity alone can be misleading. Traction Score looks for broader signs of recent attention, adoption, and active maintenance.
This skill integrates the Adaptyv Bio Foundry API and Python SDK to streamline protein engineering workflows. It enables researchers to manage the full lifecycle of laboratory experiments, from initial cost estimation and sequence submission to real-time progress monitoring.
Computational biologists can use these tools to coordinate binding screening, affinity assays, and thermostability tests. The skill provides a structured interface for interacting with target catalogs and retrieving results from enzyme activity and fluorescence workflows.
WHO IT'S FOR
Computational biologists
estimate intracellular metabolic fluxes using 13C-MFA
Protein engineers
design and monitor protein characterization experiments
Compatible AgentsThe repository documents support for these agents. The skills may also work with other agents that can load SKILL.md files, but they may need some setup or small changes.
Follow the documented setup, then try a first task.
I’m using
Interactive repository installer
Installs: Interactive selection from the Scientific Agent Skills collection; the selected skills are not fixed by this command. · Cursor, Claude Code, Codex, Google Antigravity, other compatible hosts
Before you start
A terminal environment where the documented `gh skill` command is available
Run this command in a terminal to browse and install interactively.
Research question or technical topic replacing `[research question]`
Python 3.11 or newer
Configured `EXA_API_KEY`
Internet access
Make sure exa-search is installed in your agent before trying this task.
Use the exa-search skill to perform a small scholarly first-pass search for [research question]. Bias retrieval toward publications, prioritize primary research and authoritative institutional sources, and return a concise evidence table with author names, publication year, DOI link when present, publication type, and whether peer-review status was verified or remains uncertain.
Make sure pennylane is installed in your agent before trying this task.
Use the pennylane skill to reproduce the documented one-qubit RY circuit with a Z expectation at theta = 0.3. Check its value and gradient against cosine and negative sine, perform the documented 80 gradient-descent steps, and report the final expectation value and whether each assertion passed.
A writable working directory where `vaex-example.hdf5` does not already exist
Python 3.9–3.12 with the documented Vaex dependencies
Make sure vaex is installed in your agent before trying this task.
Use the vaex skill's small executable example in a writable working directory. Create the documented synthetic table, calculate the delayed mean and count for the filtered rows, group by category, export to `vaex-example.hdf5`, reopen it, and report the checked count, mean, and schema. Stop if that output file already exists rather than replacing it.
Claude
Codex
Cursor
Gemini CLI
OpenClaw
Copilot
Kiro
Hermes
Pi
Antigravity
Paperclip