Skip to main content
A Skill is a set of instructions along with any supporting files and scripts that your AI client reads and follows from the Biohub MCP. Skills run commands from where your AI client runs them, which is typically from your own computer. Like the rest of Biohub MCP, Skills are available once your client signs in with a free Biohub account.
Your client sends your conversation, including any command output it reads, to its model provider.

Skills available through Biohub MCP

WaveOrder

WaveOrder is an open-source package from the Mehta Lab at Biohub that makes physics-based reconstructions of label-free and fluorescence microscopy data. It enables modular and user-friendly implementations of computational microscopy methods for dynamic imaging across the scales of organelles, cells, and tissues. Its source code and examples are on GitHub.

Reconstruct microscopy data

Go from a raw stack or a sample dataset to a tuned reconstruction config.

Visualize a transfer function

Compute a config’s transfer function and inspect it in napari.

TCR-Workbench

TCR-Workbench analyzes processed TCR and peptide-MHC data on your computer. It can compare peptides for an MHC, investigate TCRs, and screen T-cell repertoires. Its source code is on GitHub.

Analyze TCR data

Choose, run, and interpret a local TCR-Workbench analysis.

Set up TCR-Workbench

Install, configure, and verify a local environment for your machine.

Start a Skill

To start a Skill, use an AI client that can run commands on your computer, such as Claude Code, Codex, or Cursor. Your AI client cannot be read-only. Then, prompt your AI client with a task and the Skill name, such as “Using the Biohub MCP waveorder-tutorial Skill, …”. Your client will then read that Skill’s SKILL.md first. See the WaveOrder reconstruction tutorial or TCR-Workbench analysis to get started. See The assistant ignores a Skill for issues with getting your client to use the Skill.

Skill resources

Each Skill has a set of resources, which are files that contain instructions for your AI client to follow. Skills store these resources under the address skill://<name>/: Each Skill resource also carries the Skill’s version and every file’s size and SHA-256 hash in its MCP _meta, under biohub.ai/skill-provenance. Use that live metadata, not a value copied from these docs, when you need the exact deployed version. If your client can’t read MCP resources directly, use the list_resources and read_resource tools described in the MCP resource reference to list and read these files.