
LatchBio Workflow Integration
Build, register, debug and run bioinformatics pipelines on the Latch platform
- Repository
- K-Dense-AI/scientific-agent-skills
- Stars
- 55,780
- Licence
- MIT
- Files
- 11 files
What it does
This skill gives an AI agent detailed working knowledge of the Latch platform, a cloud service for running bioinformatics pipelines. Once installed, the agent can write Latch workflows in Python, package Nextflow or Snakemake pipelines, configure compute resources, move data in Latch Data, read Latch Registry tables, and launch or monitor runs. It routes each task to a specific reference file so it reads only what it needs.
Great for
- Bioinformatics engineers deploying or maintaining pipelines on Latch
- Developers debugging Latch registration, staging or resource configuration errors
- Teams wanting an agent to launch and monitor Latch workflow runs through the SDK or MCP
Skip it if
- Users without a Latch account or workspace, since most steps need authenticated access and will fail
- Anyone who expects the agent to validate cloud cost or results automatically: the skill warns to review cost and parameters before launching, and registration or launches change remote state
- General bioinformatics questions unrelated to the Latch platform
Try asking
- “Create a Python Latch workflow that wraps my FASTQ QC tool and register it to my workspace”
- “Configure a GPU task with 64 GB memory and retries for my alignment step”
- “Debug why my Nextflow project fails during latch register --staging”
- “Set up Latch MCP and list the workflows I can launch”
Install in 3 steps
- 1
Download the skill
The ZIP contains the 11 original files, the MIT licence and an attribution note.
latchbio-integration.zip - 2
Add it to your agent
Claude Code (all projects)
mkdir -p ~/.claude/skills && unzip latchbio-integration.zip -d ~/.claude/skills/Claude Code / Agent SDK (this project only)
mkdir -p .claude/skills && unzip latchbio-integration.zip -d .claude/skills/Claude apps (web/desktop): Settings → Capabilities → Skills → Upload skill, and choose the ZIP.
- 3
Ask for the task
No special command needed — the agent reads the skill's description and loads it when your request matches. Try one of the prompts above.
Get better results
- Run scripts/inspect_latch_sdk.py against your installed SDK version before relying on any function name, since it only performs local imports.
- Check the installed SDK version and changelog first; the guides note that some tutorials pin older Python ranges or pre-release versions.
- Review workflow parameters and expected cost in the Latch Console before launching a run, and select the workspace explicitly with latch workspace --id.
What's inside
- references/data-management.md
- references/latch-mcp.md
- references/nextflow-snakemake.md
- references/operations-and-debugging.md
- references/registry.md
- references/resource-configuration.md
- references/ui-and-automation.md
- references/verified-workflows.md
- references/workflow-creation.md
- scripts/inspect_latch_sdk.py
- SKILL.md
Redistributed unmodified from K-Dense-AI/scientific-agent-skills under the MIT licence. The assessment was drafted with AI from the skill's files on 2026-10-29. Skills can instruct your agent to run code — read SKILL.md and any scripts before installing.