latchbio-integration

Latch platform for bioinformatics workflows. Build pipelines with Latch SDK, @workflow/@task decorators, deploy serverless workflows, LatchFile/LatchDir, Nextflow/Snakemake integration.

1.65x

Quality

46%

Does it follow best practices?

Impact

91%

1.65x

Average score across 3 eval scenarios

Securityby

Passed

No known issues

Optimize this skill with Tessl

npx tessl skill review --optimize ./scientific-skills/latchbio-integration/SKILL.md

Evaluation results

81%

Multi-Sample FASTQ Processing Pipeline

Latch workflow structure and task decorators

Criteria

Without context

With context

@workflow decorator used

100%

Task decorators used

87%

100%

LatchFile type used

100%

latch:/// path format

100%

Type annotations on all params

100%

Docstrings present

100%

Right-sized QC task

100%

Right-sized alignment task

100%

map_task for batch processing

100%

LatchDir type used

100%

Correct latch imports

100%

Correct output LatchFile pattern

100%

86%

Experiment Sample Tracking System

Registry system with linked tables and bulk operations

Criteria

Without context

With context

Project import and usage

100%

Table import and creation

50%

100%

Record import and usage

100%

enum column type

100%

link column type

100%

file column type used

100%

latch:/// paths in records

100%

bulk_create for multiple records

100%

Transaction for bulk updates

100%

Transaction import

100%

Two linked tables

30%

100%

92%

18%

Protein Structure Prediction Pipeline with Custom Resource Allocation

Custom resource configuration and verified workflow integration

Criteria

Without context

With context

@small_task for lightweight steps

100%

@large_gpu_task or GPU custom_task

90%

100%

@custom_task used

100%

cpu parameter in custom_task

100%

memory parameter in custom_task

100%

gpu parameter in custom_task

50%

100%

Valid gpu_type string

100%

timeout parameter used

28%

100%

latch.verified import

100%

alphafold model_preset

colabfold mentioned as alternative

100%

Does NOT use @large_task for lightweight tasks

100%

Repository: K-Dense-AI/claude-scientific-skills
Commit: b58ad7e

Evaluated: 27 days ago
Agent: Claude Code
Model: Claude Sonnet 4.6

Table of Contents

Multi-Sample FASTQ Processing Pipeline Experiment Sample Tracking System Protein Structure Prediction Pipeline with Custom Resource Allocation

Is this your skill?

If you maintain this skill, you can claim it as your own. Once claimed, you can manage eval scenarios, bundle related skills, attach documentation or rules, and ensure cross-agent compatibility.