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
input-preparationBuild, validate, convert, and troubleshoot AlphaFold 3 input JSON files before prediction runs. Use when creating fold inputs, validating protein/RNA/DNA/ligand entries, converting AlphaFold Server JSON, adding MSAs/templates/user CCD/bonds, or explaining schema versions.
Independent GitHub snapshot · Automated classification · Original documentation governs
Use this sub-skill when the task is about constructing or checking AlphaFold 3 fold-input JSON, not about running inference or interpreting outputs.
alphafold3 dialect JSON files with name, modelSeeds, sequences, dialect, and version.userCCD, userCCDPath, and bondedAtomPairs fields.Input.from_json(...) errors before invoking a prediction run.unpairedMsaPath, pairedMsaPath, mmcifPath, and userCCDPath.../running-predictions/.../output-interpretation/.../python-apis/.dialect: "alphafold3" and version: 4 unless compatibility with an older saved input is required.id; use a list of IDs only for identical copies.unpairedMsaPath, pairedMsaPath, mmcifPath, and userCCDPath.python sub-skills/input-preparation/scripts/validate_fold_input.py fold_input.json
references/input-json.md covers schema recipes, entity fields, conversion constraints, relative paths, and validation rules.references/troubleshooting.md maps common parser and preparation failures to fixes.scripts/validate_fold_input.py provides a safe local parser check using the installed AlphaFold 3 package.