"Prepare and validate ColabFold FASTA, CSV, A3M, AlphaFold3
Scanned 9/8/2026
Install to Claude Code
npx -y skills add VectorSpaceLab/AREX-Skill --skill inputs-and-formats --agent claude-codeInstalls into .claude/skills of the current project.
Are you the author of Inputs And Formats?
Add the live security badge to your README — it updates automatically with every re-scan.
[](https://www.skillsdirectory.com/skills/vectorspacelab-inputs-and-formats)More formats (shields.io, HTML) on the badges page.
---
name: inputs-and-formats
description: "Prepare and validate ColabFold FASTA, CSV, A3M, AlphaFold3
molecule, and PDB/mmCIF inputs before MSA search or prediction."
disable-model-invocation: true
metadata:
disco-role: operating
license: MIT
---
# ColabFold inputs and formats
Use this sub-skill when the task is about making ColabFold input files valid, explaining how `get_queries` will read them, serializing A3M content, or extracting sequence/template-ready content from PDB/mmCIF files.
## Route by task
- **FASTA, CSV, TSV, or directory inputs**: use [data-formats.md](references/data-formats.md) to choose file layout, multimer syntax, sorting, and naming behavior.
- **AlphaFold3 non-protein components**: use [data-formats.md](references/data-formats.md#af3-non-protein-molecule-syntax) for `dna|`, `rna|`, `ccd|`, and `smiles|` entries, including SMILES semicolon escaping.
- **Parser/API behavior**: use [api-reference.md](references/api-reference.md) for `parse_fasta`, `classify_molecules`, `get_queries`, `msa_to_str`, and `pdb_to_string` signatures and return shapes.
- **Input validation**: run [`scripts/validate_colabfold_input.py`](scripts/validate_colabfold_input.py) before recommending expensive search or prediction.
- **Format failures**: use [troubleshooting.md](references/troubleshooting.md) for optional dependency, backend, data/config, CLI/API, and workflow failures owned by input preparation.
## Safe validation commands
```bash
python sub-skills/inputs-and-formats/scripts/validate_colabfold_input.py input.fasta
python sub-skills/inputs-and-formats/scripts/validate_colabfold_input.py input.csv --sort none
python sub-skills/inputs-and-formats/scripts/validate_colabfold_input.py msas/ --sort msa_depth --json
```
The validator only reads local input files, imports lightweight ColabFold parsing helpers when available, and performs no network, database, GPU, prediction, relaxation, or destructive actions.
## Boundaries
- For generating/searching MSAs, route to [`../msa-search/SKILL.md`](../msa-search/SKILL.md).
- For `colabfold_batch` model execution, templates during prediction, and output ranking, route to [`../batch-prediction/SKILL.md`](../batch-prediction/SKILL.md).
- For Amber relaxation, PDB/mmCIF output inspection after prediction, confidence plots, and citations, route to [`../relaxation-and-outputs/SKILL.md`](../relaxation-and-outputs/SKILL.md).
Do not send future agents back to the source repository for input syntax; this sub-skill distills the needed parser and format behavior into bundled references and scripts.

Is this your skill, or is something wrong with this listing? Request removal or report an issue. Author removals are honored within 72 hours.
No comments yet. Be the first to comment!