bioqc-mcp

Runs FastQC and MultiQC pipelines on FASTQ files and generates QC visualizations.

1.1k|257|Updated Feb 25, 2026
One-click install
npx skills add https://github.com/ClawBio/ClawBio --skill bioqc-mcp
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: bioqc-mcp
Source: https://github.com/ClawBio/ClawBio/tree/main/skills/bioqc-mcp
Command: npx skills add https://github.com/ClawBio/ClawBio --skill bioqc-mcp

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill requires matplotlib, seaborn, pandas, numpy, mcp.

What problem does it solve?

Running sequencing quality control requires manually chaining FastQC, MultiQC, log parsing, and custom plotting scripts. This Skill automates the entire workflow: it discovers FASTQ files, executes FastQC in parallel, aggregates results with MultiQC, extracts per-base quality and GC content metrics, and renders publication-quality charts, all locally.

Core Features & Use Cases

  • Automated QC Pipeline: Scans directories for FASTQ files (.fastq, .fq, .fastq.gz, .fq.gz), runs FastQC with configurable threads, and aggregates results into an interactive MultiQC HTML report.
  • Metric Extraction & Visualization: Parses fastqc_data.txt and summary.txt to extract per-base quality scores and GC content, then renders 20+ chart types (line, violin, heatmap, box plot) with Matplotlib and Seaborn.
  • Dual CLI/MCP Interface: Runs as a standard ClawBio CLI skill or as an MCP stdio server exposing 10 tools to AI agents in Cursor, Claude Desktop, and other MCP clients.
  • Use Case: A researcher receives a batch of raw sequencing reads and asks the agent to run quality control; the skill produces a pass/warn/fail summary table, quality plots, an interactive MultiQC report, and a reproducibility bundle with checksums.

Quick Start

Ask the agent to run quality control on your FASTQ directory, or execute the demo with python clawbio.py run bioqc --demo to generate a synthetic QC report with figures.

Frequently Asked Questions about bioqc-mcp

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I run FastQC and MultiQC on FASTQ files automatically?

Point the skill at a directory containing FASTQ files and it discovers all .fastq, .fq, .fastq.gz, and .fq.gz files, runs FastQC in parallel with configurable threads, then aggregates results with MultiQC into a single interactive HTML report.

How to visualize per-base quality scores and GC content from FastQC output?

The skill parses fastqc_data.txt to extract per-base sequence quality and per-sequence GC content blocks, then renders line charts, violin plots, heatmaps, and 20+ other chart types using Matplotlib and Seaborn, saved as PNG files.

Can I use FastQC quality control with Claude Desktop or Cursor?

Yes, the skill runs as an MCP stdio server exposing 10 tools including run_fastqc, run_multiqc, list_fastq_files, and generate_chart. Start it with the --mode mcp flag and register it in your MCP client configuration.

Why does the QC pipeline fail with FastQC not found?

The pipeline requires fastqc and multiqc binaries on your system PATH. Install FastQC via your package manager (e.g., brew install fastqc) and MultiQC via pip install multiqc, then rerun the pipeline.

Does sequencing QC data get uploaded to external servers?

No, all FastQC and MultiQC processing runs strictly locally via subprocess calls with no shell interpolation. No genetic data is uploaded anywhere, and every report includes a research-use disclaimer.