メニュー
Generate synthetic training data, manage seeds, preview synthesis jobs, and work with the data factory. Use when asked to generate synthetic data, create training examples, augment a dataset, manage seeds, or check factory job status.
デフォルトでは、最初にソースを確認する Prompt が選択されています。直接コマンドに切り替えるか、ローカルコピーをダウンロードすることもできます。
インストールを決める前に、SKILL.md と SkillsMP に表示されている付属ファイルをお読みください。
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。
直接コマンドでは確認用 Prompt が省略されます。実行前にソースを確認してください。
npx skills add https://github.com/GhostPeony/bashgym --skill factoryコマンドは1行のまま表示されます。コピー前に横へスクロールして全体を確認してください。
ローカルで確認しますか？SkillsMP が現在取得できるファイルをダウンロードできます。
SKILL.md を表示中
SOC 職業分類に基づく
| name | factory |
| description | Generate synthetic training data, manage seeds, preview synthesis jobs, and work with the data factory. Use when asked to generate synthetic data, create training examples, augment a dataset, manage seeds, or check factory job status. |
Synthetic data generation and training data management.
List existing seeds:
bashgym api GET /api/factory/seeds
Create a new seed:
Save {"data":{"content":"..."},"source":"manual"} as seed-request.json, then run:
bashgym api POST /api/factory/seeds --data-file seed-request.json
Create seeds from gold traces:
Save {} as empty-request.json, then run:
bashgym api POST /api/factory/seeds/from-traces --data-file empty-request.json
Delete a seed:
bashgym api DELETE /api/factory/seeds/{seed_id}
Preview what synthesis would produce:
Save {"row_count":50} as preview-request.json, then run:
bashgym api POST /api/factory/preview --data-file preview-request.json
Start synthesis job:
Save {"row_count":10} as synthesis-request.json, then run:
bashgym api POST /api/factory/synthesize --data-file synthesis-request.json
Generate synthetic training examples:
Save {"preset":"custom","target_examples":50} as
synthetic-request.json, then run:
bashgym api POST /api/factory/synthetic/generate --data-file synthetic-request.json
List synthetic generation jobs:
bashgym api GET /api/factory/synthetic/jobs
Check job status:
bashgym api GET /api/factory/synthetic/jobs/{job_id}
Available presets:
bashgym api GET /api/factory/synthetic/presets
Read the current full configuration, edit that schema into factory-config.json,
then update it:
bashgym api GET /api/factory/config
bashgym api PUT /api/factory/config --data-file factory-config.json
User: "Generate 50 synthetic training examples"
→ Call bashgym api GET /api/factory/synthetic/presets to show options. Ask which preset.
Call bashgym api POST /api/factory/synthetic/generate --data-file synthetic-request.json. Report job_id.
User: "Create seeds from my gold traces"
→ Call bashgym api POST /api/factory/seeds/from-traces --data-file empty-request.json. Report how many seeds created.
User: "What seeds do I have?"
→ Call bashgym api GET /api/factory/seeds. List source and data previews.
Operate BashGym training, evaluation, and durable AutoResearch from Codex, Claude Code, Hermes, or another compatible agent. Use for guided campaign preparation and resumption, ML session planning, preflight, goals and KPIs, model training or fine-tuning, runtime monitoring, evaluation and benchmark comparison, report generation, bounded iteration, and curated GBrain updates across embedding, SFT, DPO, GRPO/RLVR, LoRA/QLoRA, distillation, and full-model workflows.
Browse, compare, evaluate, and manage trained models. View leaderboard, trends, lineage, and artifacts. Deploy models to Ollama. Use when asked about models, model comparison, which model is best, model evaluation, deploying a model, or downloading a model.
Check BashGym system status, GPU utilization, hardware info, and aggregate statistics. Use when asked about system health, what's running, GPU usage, memory, disk space, or general system stats.