MFA Pipeline Orchestrator
Runs a full metabolic flux analysis pipeline (model to yield and figures) by coordinating sub-agents.
- Repository
- aiming-lab/AutoResearchClaw
- Stars
- 14,802
- Licence
- MIT
- Files
- 1 file
What it does
This skill teaches an AI agent to manage a multi-step computational biology workflow: loading a genome-scale metabolic model (from a BIGG model ID or a custom reaction list), simulating growth and metabolite production, analysing which genes are essential, and producing publication-style figures and a yield table. It does not do the maths itself; it hands each stage to specialised sub-agents and records progress in markdown files so a failed step can be resumed rather than restarted.
Great for
- Researchers or students in metabolic engineering who want an automated first pass at flux balance analysis on a known organism
- Computational biologists who already work with COBRApy and BIGG models and want to chain several analyses together
- Teams exploring knockout targets or theoretical yield for a product before lab work
Skip it if
- Clinical or medical decisions: the outputs are model predictions from simplified assumptions and should not guide treatment or safety-critical choices
- Users without a working understanding of metabolic models, since the results need expert interpretation and the skill does not validate biological accuracy
- Situations needing a single quick calculation, because the multi-agent pipeline adds overhead the user does not need
Try asking
- “Run an end-to-end flux analysis of E. coli core model iJO1366 with glucose as carbon source and ethanol as the target product.”
- “Compare wild-type and gene-knockout growth for a custom reaction list and report essential genes.”
- “Generate a phase plane plot for glucose and oxygen uptake and list the top engineering targets for succinate yield.”
Install in 3 steps
- 1
Download the skill
The ZIP contains the 1 original files, the MIT licence and an attribution note.
mfa-pipeline-orchestrator.zip - 2
Add it to your agent
Claude Code (all projects)
mkdir -p ~/.claude/skills && unzip mfa-pipeline-orchestrator.zip -d ~/.claude/skills/Claude Code / Agent SDK (this project only)
mkdir -p .claude/skills && unzip mfa-pipeline-orchestrator.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
- Provide the BIGG model ID, the carbon source and concentration, the oxygen condition and the target product up front so Step 0 captures complete inputs.
- Check the progress/ files after a run; each step reports PASS or FAIL, which tells you where to resume.
- Treat the yield and essential-gene outputs as hypotheses to verify with the original model documentation or wet-lab experiments.
What's inside
- SKILL.md
Redistributed unmodified from aiming-lab/AutoResearchClaw 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.