executing-plans
executing-plans
Use when executing an implementation plan in the current session as the implementer yourself — your human partner chose inline execution, or no subagent tool is available
inputs-and-formatsPrepare and validate ColabFold FASTA, CSV, A3M, AlphaFold3 molecule, and PDB/mmCIF inputs before MSA search or prediction.
Independent GitHub snapshot · Automated classification · Original documentation governs
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.
dna|, rna|, ccd|, and smiles| entries, including SMILES semicolon escaping.parse_fasta, classify_molecules, get_queries, msa_to_str, and pdb_to_string signatures and return shapes.scripts/validate_colabfold_input.py before recommending expensive search or prediction.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.
../msa-search/SKILL.md.colabfold_batch model execution, templates during prediction, and output ranking, route to ../batch-prediction/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.