sushi-framework

Submit, monitor, and administer SUSHI bioinformatics jobs on FGCZ SLURM infrastructure.

Updated May 28, 2026
One-click install
npx skills add https://github.com/cpanse/skills --skill sushi-framework-cpanse
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: sushi-framework
Source: https://github.com/cpanse/skills/tree/main/fgcz-infrastructure/skills/sushi-framework
Command: npx skills add https://github.com/cpanse/skills --skill sushi-framework-cpanse

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes scripts (resource) and references (resource) components.

What problem does it solve? Operating the SUSHI framework at FGCZ requires knowing web workflows, CLI syntax, server deployment steps, and debugging procedures scattered across systems. This Skill consolidates user operations, admin deployment, and troubleshooting knowledge for SUSHI into one operational guide. ## Core Features & Use Cases - Job Submission & Monitoring: Submit analysis jobs via the SUSHI web interface or the sushi_fabric CLI, create dataset and parameterset TSV files, and track job status through SLURM. - Server Administration: Deploy production, demo, course, and test SUSHI instances, manage the Python Job Manager daemon, and administer the MySQL database. - Troubleshooting & Integration: Diagnose failed jobs, B-Fabric registration errors, and gStore copy issues using documented log locations and recovery procedures. - Use Case: A bioinformatician needs to run FastQC on a new RNA-seq dataset for project p1535. The Skill provides the exact sushi_fabric command, dataset TSV format, and monitoring commands to submit and verify the job. ## Quick Start Ask the agent to submit a FastQC job to SUSHI for project 1535 using a dataset TSV file and show how to monitor its status.

Frequently Asked Questions about sushi-framework

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

FAQPage Schema
How do I submit a SUSHI job from the command line?

Use the sushi_fabric command from /srv/sushi/production/master with bundle exec. Provide the app class with --class, a dataset TSV with --dataset, the project number, and add --run to execute; omitting --run performs a validation-only test.

What format does a SUSHI dataset TSV file require?

Dataset files are tab-separated with a Name column plus typed columns like Read1 [File] for file paths, Report [Link] for URLs, and Condition [Factor] for experimental factors. Each app defines required columns, such as Name, Read1, and Species for STARApp.

How do I set up a test SUSHI instance for development?

SSH to fgcz-h-083, load the Ruby 3.3.7 module and conda environment, then run setup_test_sushi_latest.sh from /srv/sushi. Start the server with bundle exec rails s on port 5000 and access it via FGCZ VPN.

Why are my SUSHI jobs stuck in pending status?

Pending jobs usually mean the Job Manager daemon is not running or cannot reach the MySQL jobs table. Check the daemon lock file at /misc/fgcz01/sushi/sushi_jobmanager_daemon.lock, review its log, and restart it with start_sushi_jobmanager.py.

Does SUSHI integrate with B-Fabric LIMS?

Yes, SUSHI uses B-Fabric for LDAP authentication, project access control, and dataset registration. Output datasets are registered via the run_dataset_register.rb daemon, and registration can be disabled per submission with the --off_bfab flag.

Why does a SAMPLE mode SUSHI app fail with a nil error in next_dataset?

During validation, sushi_fabric resets @dataset to an empty hash before calling next_dataset, so @dataset['Name'] returns nil. Add a fallback like @dataset['Name'] || 'placeholder' in next_dataset, then restart Apache to reload the code.