跳轉到

AI/ML Research —— AI/ML 研究

Terminology rule (zh-TW pages)

技術名詞首次出現以「中文 (English original)」格式呈現，例：依賴注入 (dependency injection)。不自創翻譯——若無公認譯名直接保留英文 （如 embedding、tokenizer）。代碼、API 名、CLI flag、套件名、檔名一律不翻。

實驗追蹤 (experiment tracking)、模型生命週期 (model lifecycle)、 資料版本管理 (data versioning)、notebook、fine-tuning、agent framework 與更廣的 research-engineering 工具 stack。

此 repo 內的 skill

Local

Skill 一句話 備註
mlflow-tracking 通用 MLflow —— 單機 sqlite + mlflow ui、團隊用 PostgreSQL + MinIO docker；附 LLM tracing、registry、autolog。
dvc-ml-workflow DVC pipeline + queued experiment，metrics 自動綁到臨時 commit。
marimo-batch-mlflow marimo 雙模式（UI + batch CLI）notebook，搭配 Tyro + MLflow。

Vendored

Skill Upstream Series
marimo-notebook marimo-team/skills flat
streamlit-to-marimo marimo-team/skills flat
anywidget marimo-team/skills flat
deep-research 199-biotechnologies/deep-research deep-research series

External skills（手動安裝）

Status enum (single source of truth across the catalog)

Every external-skill or MCP entry below carries a status: value:

Status Meaning Where it links
vendored Already in vendor.yaml. Link to the skills/vendor/<name>/ (or series) entry.
deferred Open TODO P? item — under consideration. Link to the TODO.md anchor.
skipped Looked at, chose not to vendor. Inline reason required (1 sentence).
evaluated Read but no decision recorded. Effectively "tracked." Inline 1-line note describing what was learned.
wishlist Surfaced but not yet evaluated. No link required; default for fresh discoveries.

Status changes are explicit edits — see the Adding catalog entries workflow doc (under Workflows in the nav) for the wishlist → deferred → vendored recipe.

Manual install for non-vendored skills

When a skill is not in vendor.yaml, you can still install it directly from upstream via the npx skills CLI:

# Whole-repo install (if SKILL.md sits at repo root or skills/)
npx skills@latest add <owner>/<repo>

# Subpath install (when the skill is nested)
npx skills@latest add <owner>/<repo>/<path-to-skill-dir>

# Cherry-pick by name
npx skills@latest add <owner>/<repo> -s <skill-name>

# Force deep discovery (skill nested >1 level inside skills/)
npx skills@latest add <owner>/<repo> --full-depth -s <skill-name>

For Anthropic-style plugin marketplaces (e.g. anthropics/financial-services, anthropics/knowledge-work-plugins), use the Claude Code plugin commands instead:

claude plugin marketplace add <owner>/<repo>
claude plugin install <plugin-name>@<repo-name>

For MCP servers, see the per-entry "Auth & install" section in the MCP wiki — install commands vary by transport (HTTP / stdio / SSE) and host (Claude Code / Claude Desktop / Cursor / Managed Agents).

Skill / Plugin Upstream Status 為何此狀態 安裝提示
AI-research-SKILLs（完整 library） Orchestra-Research/AI-research-SKILLs wishlist 跨 23 個分類的 98 個 skill —— 從 architecture（LitGPT、Mamba、NanoGPT）、fine-tuning（Axolotl、LLaMA-Factory、PEFT）到 inference（vLLM、TensorRT-LLM）的完整研究生命週期。範圍太廣不會整批 vendor，會精選 (cherry-pick)。 npx @orchestra-research/ai-research-skills（他們的 npm wrapper）或瀏覽後 npx skills@latest add Orchestra-Research/AI-research-SKILLs/<category>/<skill>。
bio-research plugin anthropics/knowledge-work-plugins/bio-research wishlist PubMed、BioRender、bioRxiv、ChEMBL、Benchling、Open Targets connector。本身不直接適用，但作為 life-sciences research workflow 的參考 (reference) 有價值。 claude plugin install bio-research@knowledge-work-plugins

MCP servers

名稱 Upstream Status Auth 紀錄
尚未調查

Backlog（TODO P? 條目）

見 TODO.md 的 P? 區：

  • [?/L] LangChain / LangSmith / LangServe / LangGraph / Langfuse —— 把 orchestration + observability stack 包成一組 skill。
  • [?/L] LLM fine-tuning skill —— 監督式 (supervised) fine-tuning 與 adapter-based tuning 的實作 workflow。
  • [?/M] Build MCP skill —— 建立、測試、記錄 MCP server 的最小 workflow（注意：mcp-builder 已 vendored —— 若已涵蓋則重新評估）。
  • [?/M] Hugging Face Spaces + Gradio skill —— demo app 部署、 secret 處理、本機到託管 (hosted) 的交接。
  • [?/M] Data visualization skill —— Matplotlib + Seaborn + Plotly 用於探索性 (exploratory) + 報告級 (report-ready) 圖表。

另見