メニュー
Run ColabFold MSA search from a FASTA file and automatically clean resulting A3M files into paired/unpaired outputs for downstream AF3-style pipelines.
Codex または Claude でインストール この Prompt をコピーして Codex、Claude、または他のアシスタントに貼り付けると、Skill ページを確認してインストールできます。
SOC 職業分類に基づく
Perform normalized filesystem operations (list, read, write, append, mkdir, stat, copy, move, delete, search, replace) through a single skill script. Use when tasks require deterministic file manipulation through get_skill_script instead of built-in FileTools.
Execute Python snippets or Python files through a normalized skill entrypoint with explicit cwd, timeout, and argv. Use when tasks require ad-hoc Python execution and you want skill-based execution instead of built-in PythonTools.
Execute shell commands through a normalized skill entrypoint with explicit cwd and timeout. Use when tasks require command-line operations (tests, build, git, search, file inspection) and you want script-based execution via get_skill_script instead of built-in ShellTools.
Convert AlphaFold3 official inference JSON into Protenix, Boltz, or OpenFold3 inputs with schema validation. Use when an existing AF3 JSON must be reused by another folding stack, including batch conversion, target auto-detection from output filename, and best-effort mapping of AF3 `bondedAtomPairs`, MSA fields, templates, and ligand CCD/smiles data.
Convert CSV tables into valid AlphaFold3 input JSONs. Use this skill when you need to batch-generate AF3 JSONs from tabular inputs and run in MSA-free mode (protein uses empty unpairedMsa/pairedMsa, RNA uses empty unpairedMsa).
Extract deduplicated protein, DNA, and RNA sequences from AF3 or Protenix inference JSON files into per-entity-type FASTA files keyed by SHA-256 hash. Use as the first stage of the MSA pipeline before `af3-msa-search` and `af3-msa-injector`.
| name | af3-msa-search |
| description | Run ColabFold MSA search from a FASTA file and automatically clean resulting A3M files into paired/unpaired outputs for downstream AF3-style pipelines. |
Generate MSA artifacts from one FASTA file, then clean them:
.a3m files under msas/unpairedMSA/ and pairedMSA/ under processed_msas/.done markersUse this single entrypoint:
bash skills/af3-msa-search/scripts/start_production.sh \
--query /abs/path/to/all.fasta
start_production.sh behavior:
tmux session (SSH-safe),launch_and_watch.sh immediately,launch_and_watch.sh starts run.sh (inner tmux by default),${out_root}/.ready_for_next_step when finished.No next-step handoff hooks are included in this skill.
When using this skill in an interactive chat, do not wait for full completion.
After launching start_production.sh, perform only a minimal health check, then end the turn:
tmux has-session -t <session>)[Split] / [Run] batch_...)[Error], [Fatal], Traceback) in pipeline.log and logs/If checks pass, report the session/log paths and stop. Let the job continue in background.
This skill reads only LOCALCOLABFOLD_* variables from the current shell.
If you store them in .envrc, load/export them before running.
Common variables:
LOCALCOLABFOLD_IMAGELOCALCOLABFOLD_DB_DIRLOCALCOLABFOLD_GPUSLOCALCOLABFOLD_THREADSLOCALCOLABFOLD_BATCH_SIZELOCALCOLABFOLD_RESUMELOCALCOLABFOLD_USE_GPU_SERVERLOCALCOLABFOLD_QUERY, LOCALCOLABFOLD_OUT_ROOT, LOCALCOLABFOLD_TMUX_SESSIONUnder --out-root:
msas/batch_000001/ : raw MSA files (*.a3m) and .doneprocessed_msas/batch_000001/unpairedMSA/ : cleaned unpaired outputsprocessed_msas/batch_000001/pairedMSA/ : cleaned paired outputslogs/ : gpuserver logs + per-batch logstmp/batches/ : split FASTA chunks.ready_for_next_step : completion marker after cleanupIf --resume 1, completed batches are skipped.
After completion (offline follow-up):
logs/batch_*.logmsas/<batch>/.doneprocessed_msas/${db}.dbtype and ${db}.idx/${db}.index.a3m: inspect corresponding logs/batch_*.loglogs/gpuserver_*.log, retry with --use-gpu-server 0