Metadata-Version: 2.4
Name: data-science-mcp
Version: 1.0.1
Summary: Data Science MCP Server — Model training, evaluation, and evolution tools for agentic ML workflows. Integrates with agent-utilities IModelEvolver (CONCEPT:AHE-3.15).
Author-email: Audel Rouhi <knucklessg1@gmail.com>
License: MIT
Classifier: Development Status :: 4 - Beta
Classifier: License :: OSI Approved :: MIT License
Classifier: Environment :: Console
Classifier: Operating System :: POSIX :: Linux
Classifier: Programming Language :: Python :: 3
Requires-Python: <3.14,>=3.11
Description-Content-Type: text/markdown
License-File: LICENSE
Requires-Dist: agent-utilities[mcp]>=1.0.0
Requires-Dist: epistemic-graph[datascience]>=2.0.0
Requires-Dist: polars>=1.0.0
Requires-Dist: numpy>=1.24.0
Provides-Extra: mcp
Requires-Dist: agent-utilities[mcp]>=1.0.0; extra == "mcp"
Provides-Extra: agent
Requires-Dist: agent-utilities[agent,logfire]>=1.0.0; extra == "agent"
Provides-Extra: datasets
Requires-Dist: scikit-learn>=1.5.0; extra == "datasets"
Provides-Extra: training
Requires-Dist: torch>=2.4.0; extra == "training"
Requires-Dist: transformers>=4.44.0; extra == "training"
Requires-Dist: peft>=0.12.0; extra == "training"
Requires-Dist: bitsandbytes>=0.43.0; extra == "training"
Requires-Dist: accelerate>=0.34.0; extra == "training"
Requires-Dist: datasets>=2.20.0; extra == "training"
Requires-Dist: tokenizers>=0.19.0; extra == "training"
Requires-Dist: httpx>=0.27.0; extra == "training"
Requires-Dist: scikit-learn>=1.5.0; extra == "training"
Provides-Extra: training-scale
Requires-Dist: deepspeed>=0.14.0; extra == "training-scale"
Requires-Dist: flash-attn>=2.6.0; extra == "training-scale"
Provides-Extra: training-fast
Requires-Dist: liger-kernel>=0.3.0; extra == "training-fast"
Provides-Extra: eval
Requires-Dist: lm-eval>=0.4.0; extra == "eval"
Provides-Extra: tracking
Requires-Dist: mlflow>=2.14.0; extra == "tracking"
Provides-Extra: all
Requires-Dist: agent-utilities[agent,logfire,mcp]>=1.0.0; extra == "all"
Requires-Dist: epistemic-graph[datascience]>=2.0.0; extra == "all"
Requires-Dist: scikit-learn>=1.5.0; extra == "all"
Dynamic: license-file

# Data Science Mcp
## CLI or API | MCP | Agent

![PyPI - Version](https://img.shields.io/pypi/v/data-science-mcp)
![MCP Server](https://badge.mcpx.dev?type=server 'MCP Server')
![PyPI - Downloads](https://img.shields.io/pypi/dd/data-science-mcp)
![GitHub Repo stars](https://img.shields.io/github/stars/Knuckles-Team/data-science-mcp)
![GitHub forks](https://img.shields.io/github/forks/Knuckles-Team/data-science-mcp)
![GitHub contributors](https://img.shields.io/github/contributors/Knuckles-Team/data-science-mcp)
![PyPI - License](https://img.shields.io/pypi/l/data-science-mcp)
![GitHub](https://img.shields.io/github/license/Knuckles-Team/data-science-mcp)
![GitHub last commit (by committer)](https://img.shields.io/github/last-commit/Knuckles-Team/data-science-mcp)
![GitHub pull requests](https://img.shields.io/github/issues-pr/Knuckles-Team/data-science-mcp)
![GitHub closed pull requests](https://img.shields.io/github/issues-pr-closed/Knuckles-Team/data-science-mcp)
![GitHub issues](https://img.shields.io/github/issues/Knuckles-Team/data-science-mcp)
![GitHub top language](https://img.shields.io/github/languages/top/Knuckles-Team/data-science-mcp)
![GitHub language count](https://img.shields.io/github/languages/count/Knuckles-Team/data-science-mcp)
![GitHub repo size](https://img.shields.io/github/repo-size/Knuckles-Team/data-science-mcp)
![GitHub repo file count (file type)](https://img.shields.io/github/directory-file-count/Knuckles-Team/data-science-mcp)
![PyPI - Wheel](https://img.shields.io/pypi/wheel/data-science-mcp)
![PyPI - Implementation](https://img.shields.io/pypi/implementation/data-science-mcp)

*Version: 1.0.1*

> **Documentation** — Installation, deployment, usage across the MCP, Python API, and
> CLI interfaces, and the in-house model-training substrate are maintained in the
> [official documentation](https://knuckles-team.github.io/data-science-mcp/).

---

## Overview

**Data Science Mcp** is a production-grade Agent and Model Context Protocol (MCP) server designed to interface directly with Data Science MCP Server — Model training, evaluation, and evolution tools for agentic ML workflows. Integrates with agent-utilities IModelEvolver (CONCEPT:AHE-3.8)..

---

## Key Features

- **Consolidated Action-Routed MCP Tools:** Minimizes token overhead and eliminates tool bloat in LLM contexts by grouping methods into optimized, togglable tool modules.
- **Enterprise-Grade Security:** Comprehensive support for Eunomia policies, OIDC token delegation, and granular execution context tracking.
- **Integrated Graph Agent:** Built-in Pydantic AI agent supporting the Agent Control Protocol (ACP) and standard Web interfaces (AG-UI).
- **Native Telemetry & Tracing:** Out-of-the-box OpenTelemetry exports and native Langfuse tracing.
- **In-House Model Training (Wave C):** A deterministic SFT/DPO/GRPO corpus + reward engine plus torch/PEFT gradient trainers (LoRA/QLoRA, TIES adapter merge, vLLM rollouts, checkpoint→reliability-suite eval hooks). Loss/optimizer kernels are CPU-smoke-tested on a toy model; real fine-tunes run on the GB10. Install with `pip install data-science-mcp[training]`. MCP tools: `build_training_dataset`, `compose_reward`, `train_sft`, `train_dpo`, `train_grpo`, `merge_adapters_ties`. See **[docs/training.md](docs/training.md)**.

---

## CLI or API

This agent wraps the Data Science MCP Server — Model training, evaluation, and evolution tools for agentic ML workflows. Integrates with agent-utilities IModelEvolver (CONCEPT:AHE-3.8). API. You can interact with it programmatically or via its integrated execution entrypoints.

Detailed instructions on how to use the underlying API wrappers, extended schema bindings, and developer SDK references are maintained in [docs/index.md](docs/index.md).

---

## MCP

This server utilizes dynamic Action-Routed tools to optimize token overhead and maximize IDE compatibility.

### Available MCP Tools

_Auto-generated — do not edit (synced by the `mcp-readme-table` pre-commit hook)._

<!-- MCP-TOOLS-TABLE:START -->

#### Condensed action-routed tools (default — `MCP_TOOL_MODE=condensed`)

| MCP Tool | Toggle Env Var | Description |
|----------|----------------|-------------|
| `build_training_dataset` | `MODEL_TRAININGTOOL` | Build an SFT/DPO/GRPO training corpus from traces (CONCEPT:AHE-3.1). |
| `compose_reward` | `MODEL_TRAININGTOOL` | Composite, conditionally-gated reward score (CONCEPT:AHE-3.1). |
| `cross_validate` | `MODEL_TRAININGTOOL` | Perform k-fold cross-validation for a model class. |
| `curate_corpus` | `DATA_ENGINETOOL` | Full curation pass: quality-filter → dedup → decontaminate → lineage. |
| `dataset_lineage` | `DATA_ENGINETOOL` | Record a ``DatasetVersion`` provenance node (CONCEPT:ML-002). |
| `decontaminate_corpus` | `DATA_ENGINETOOL` | Drop training records that leak held-out eval examples (CONCEPT:ML-002). |
| `dedup_corpus` | `DATA_ENGINETOOL` | Remove exact + near-duplicate records (CONCEPT:ML-002). |
| `describe_dataset` | `DATA_MANAGEMENTTOOL` | Get descriptive statistics for a loaded dataset. |
| `ds_specialize_kernel` | `MODEL_TRAININGTOOL` | Run a SAI-factory specialization cycle on a compute kernel (CONCEPT:AHE-3.29). |
| `evaluate_model` | `MODEL_TRAININGTOOL` | Evaluate a fitted model on a dataset split. |
| `evolve_model_class` | `MODEL_EVOLUTIONTOOL` | Submit a model to the evolutionary Pareto frontier. |
| `fit_model` | `MODEL_TRAININGTOOL` | Fit a machine learning model on a dataset and return metrics. |
| `generate_interpretability_tests` | `INTERPRETABILITYTOOL` | Generate a structured suite of 6 interpretability test cases for a model. |
| `get_pareto_frontier` | `MODEL_EVOLUTIONTOOL` | Retrieve the current Pareto frontier of model classes. |
| `grade_response` | `INTERPRETABILITYTOOL` | Grade a model interpretability response against reference answer. |
| `load_dataset` | `DATA_MANAGEMENTTOOL` | Load and parse a dataset by name or CSV file path. |
| `merge_adapters_ties` | `MODEL_TRAININGTOOL` | TIES-merge multiple task vectors onto a base (MeMo; CONCEPT:AHE-3.1). |
| `predict` | `MODEL_TRAININGTOOL` | Generate predictions using a fitted model. |
| `prepare_pretrain_data` | `DATA_ENGINETOOL` | Tokenize a corpus into a flat-token HDF5 file for pretraining (CONCEPT:ML-010). |
| `pretrain_model` | `MODEL_TRAININGTOOL` | Pretrain a causal LM **from random init** (CONCEPT:ML-003). |
| `quant_derivatives` | `QUANTTOOL` | SABR stochastic-volatility surface kernels (CONCEPT:KG-2.20j). |
| `quant_forensic` | `QUANTTOOL` | Forensic-accounting report (CONCEPT:KG-2.20g). |
| `quant_market_making` | `QUANTTOOL` | Market-making / HFT quoting kernels (CONCEPT:KG-2.20f). |
| `quant_microstructure` | `QUANTTOOL` | Order-flow / toxicity / self-excitation kernels (CONCEPT:KG-2.20f). |
| `quant_signals` | `QUANTTOOL` | Signal-combination / breadth kernels (CONCEPT:KG-2.20i). |
| `quant_sizing` | `QUANTTOOL` | Position-sizing kernels (CONCEPT:KG-2.20f / KG-2.20i). |
| `quant_statespace` | `QUANTTOOL` | State-space / statistical-arbitrage kernels (CONCEPT:KG-2.20h). |
| `quant_validation` | `QUANTTOOL` | Backtest-validation / calibration kernels (CONCEPT:KG-2.20f / KG-2.20i). |
| `rank_models` | `MODEL_EVOLUTIONTOOL` | Rank all registered fitted models by their test R2 score. |
| `run_interpretability_suite` | `INTERPRETABILITYTOOL` | Run and grade the complete 6-category interpretability audit suite for a model. |
| `split_dataset` | `DATA_MANAGEMENTTOOL` | Split a loaded dataset into train, test, and validation sets. |
| `train_dpo` | `MODEL_TRAININGTOOL` | Preference-optimise on a ``dpo`` corpus (CONCEPT:AHE-3.1). |
| `train_grpo` | `MODEL_TRAININGTOOL` | GRPO on advantage-tagged groups (CONCEPT:AHE-3.1). |
| `train_ppo` | `MODEL_TRAININGTOOL` | Proximal Policy Optimization with GAE + value head (CONCEPT:ML-009). |
| `train_reward` | `MODEL_TRAININGTOOL` | Train a Bradley-Terry reward model on preference pairs (CONCEPT:ML-008). |
| `train_sft` | `MODEL_TRAININGTOOL` | Supervised fine-tune on an ``sft`` corpus (CONCEPT:AHE-3.1). |
| `train_tokenizer` | `MODEL_TRAININGTOOL` | Train a byte-level BPE tokenizer from scratch (CONCEPT:ML-003). |

#### Verbose 1:1 API-mapped tools (`MCP_TOOL_MODE=verbose` or `both`)

<details>
<summary>9 per-operation tools — one per public API method (click to expand)</summary>

| MCP Tool | Toggle Env Var | Description |
|----------|----------------|-------------|
| `data_science_cross_validate` | `ML_ENGINETOOL` | Run k-fold cross-validation via the engine. |
| `data_science_describe_dataset` | `ML_ENGINETOOL` | Get descriptive statistics for a loaded dataset. |
| `data_science_evaluate` | `ML_ENGINETOOL` | Evaluate a fitted model. |
| `data_science_fit` | `ML_ENGINETOOL` | Fit a model on a dataset via the epistemic-graph engine. |
| `data_science_interpretability_reference` | `ML_ENGINETOOL` | Compute reference answers for the interpretability suite without any |
| `data_science_load_dataset` | `ML_ENGINETOOL` | Load a dataset by name or file path. |
| `data_science_predict` | `ML_ENGINETOOL` | Generate predictions from a fitted model. |
| `data_science_ranked_models` | `ML_ENGINETOOL` | Rank fitted models by stored test R² (backend-agnostic, no recompute). |
| `data_science_split_dataset` | `ML_ENGINETOOL` | Split a dataset into train/test/validation sizes. |

</details>

_37 action-routed tool(s) (default) · 9 verbose 1:1 tool(s). Each is enabled unless its `<DOMAIN>TOOL` toggle is set false; `MCP_TOOL_MODE` selects the surface (`condensed` default · `verbose` 1:1 · `both`). Auto-generated — do not edit._
<!-- MCP-TOOLS-TABLE:END -->

Detailed tool schemas, parameter shapes, and validation constraints are preserved in [docs/mcp.md](docs/mcp.md).

### Dynamic Tool Selection & Visibility

This MCP server supports dynamic toolset selection and visibility filtering at runtime. This allows you to restrict the set of exposed tools in order to prevent blowing up the LLM's context window.

You can configure tool filtering via multiple input channels:

- **CLI Arguments:** Pass `--tools` or `--toolsets` (or their disabled counterparts `--disabled-tools` and `--disabled-toolsets`) during startup.
- **Environment Variables:** Define standard environment variables:
  - `MCP_ENABLED_TOOLS` / `MCP_DISABLED_TOOLS`
  - `MCP_ENABLED_TAGS` / `MCP_DISABLED_TAGS`
- **HTTP SSE Request Headers:** Pass custom headers during transport initialization:
  - `x-mcp-enabled-tools` / `x-mcp-disabled-tools`
  - `x-mcp-enabled-tags` / `x-mcp-disabled-tags`
- **HTTP SSE Request Query Parameters:** Append query parameters directly to your transport connection URL:
  - `?tools=tool1,tool2`
  - `?tags=tag1`

When query strings or parameters are supplied, an LLM-free **Knowledge Graph resolution layer** (using `DynamicToolOrchestrator`) matches query intents against known tool tags, names, or descriptions, with safe fallback and automated 24-hour background cache refreshing.

---

### MCP Configuration Examples

<!-- MCP-CONFIG-EXAMPLES:START -->

> **Install the slim `[mcp]` extra.** All examples install `data-science-mcp[mcp]` — the
> MCP-server extra that pulls only the FastMCP / FastAPI tooling (`agent-utilities[mcp]`).
> It deliberately **excludes** the heavy agent runtime (`pydantic-ai`, the epistemic-graph
> engine, `dspy`, `llama-index`), so `uvx` / container installs are far smaller. Use the
> full `[agent]` extra only when you need the integrated Pydantic AI agent.

#### stdio Transport (local IDEs — Cursor, Claude Desktop, VS Code)

```json
{
  "mcpServers": {
    "data-science-mcp": {
      "command": "uvx",
      "args": [
        "--from",
        "data-science-mcp[mcp]",
        "data-science-mcp"
      ],
      "env": {
        "MCP_TOOL_MODE": "condensed",
        "DATA_ENGINETOOL": "True",
        "DATA_MANAGEMENTTOOL": "True",
        "DATA_SCIENCE_MCP_TOKEN": "your_token_here",
        "DATA_SCIENCE_MCP_URL": "http://localhost:8080",
        "DATA_SCIENCE_MCP_VERIFY": "True",
        "DSM_NEAR_PAIRS_LOCAL_MAX": "20000",
        "EPISTEMIC_GRAPH_SOCKET": "",
        "EPISTEMIC_GRAPH_TCP": "",
        "INFERENCE_API_KEY": "EMPTY",
        "INFERENCE_BACKEND": "vllm",
        "INFERENCE_BASE_URL": "",
        "INFERENCE_MODEL": "",
        "INTERPRETABILITYTOOL": "True",
        "KERNEL_SPECIALIZETOOL": "True",
        "MODEL_EVOLUTIONTOOL": "True",
        "MODEL_TRAININGTOOL": "True",
        "QUANTTOOL": "True",
        "TRAINERTOOL": "True",
        "TRAINING_DATATOOL": "True"
      }
    }
  }
}
```

#### Streamable-HTTP Transport (networked / production)

```json
{
  "mcpServers": {
    "data-science-mcp": {
      "command": "uvx",
      "args": [
        "--from",
        "data-science-mcp[mcp]",
        "data-science-mcp",
        "--transport",
        "streamable-http",
        "--port",
        "8000"
      ],
      "env": {
        "TRANSPORT": "streamable-http",
        "HOST": "0.0.0.0",
        "PORT": "8000",
        "MCP_TOOL_MODE": "condensed",
        "DATA_ENGINETOOL": "True",
        "DATA_MANAGEMENTTOOL": "True",
        "DATA_SCIENCE_MCP_TOKEN": "your_token_here",
        "DATA_SCIENCE_MCP_URL": "http://localhost:8080",
        "DATA_SCIENCE_MCP_VERIFY": "True",
        "DSM_NEAR_PAIRS_LOCAL_MAX": "20000",
        "EPISTEMIC_GRAPH_SOCKET": "",
        "EPISTEMIC_GRAPH_TCP": "",
        "INFERENCE_API_KEY": "EMPTY",
        "INFERENCE_BACKEND": "vllm",
        "INFERENCE_BASE_URL": "",
        "INFERENCE_MODEL": "",
        "INTERPRETABILITYTOOL": "True",
        "KERNEL_SPECIALIZETOOL": "True",
        "MODEL_EVOLUTIONTOOL": "True",
        "MODEL_TRAININGTOOL": "True",
        "QUANTTOOL": "True",
        "TRAINERTOOL": "True",
        "TRAINING_DATATOOL": "True"
      }
    }
  }
}
```

Alternatively, connect to a pre-deployed Streamable-HTTP instance by `url`:

```json
{
  "mcpServers": {
    "data-science-mcp": {
      "url": "http://localhost:8000/data-science-mcp/mcp"
    }
  }
}
```

Deploying the Streamable-HTTP server via Docker:

```bash
docker run -d \
  --name data-science-mcp-mcp \
  -p 8000:8000 \
  -e TRANSPORT=streamable-http \
  -e HOST=0.0.0.0 \
  -e PORT=8000 \
  -e MCP_TOOL_MODE=condensed \
  -e DATA_ENGINETOOL=True \
  -e DATA_MANAGEMENTTOOL=True \
  -e DATA_SCIENCE_MCP_TOKEN=your_token_here \
  -e DATA_SCIENCE_MCP_URL=http://localhost:8080 \
  -e DATA_SCIENCE_MCP_VERIFY=True \
  -e DSM_NEAR_PAIRS_LOCAL_MAX=20000 \
  -e EPISTEMIC_GRAPH_SOCKET="" \
  -e EPISTEMIC_GRAPH_TCP="" \
  -e INFERENCE_API_KEY=EMPTY \
  -e INFERENCE_BACKEND=vllm \
  -e INFERENCE_BASE_URL="" \
  -e INFERENCE_MODEL="" \
  -e INTERPRETABILITYTOOL=True \
  -e KERNEL_SPECIALIZETOOL=True \
  -e MODEL_EVOLUTIONTOOL=True \
  -e MODEL_TRAININGTOOL=True \
  -e QUANTTOOL=True \
  -e TRAINERTOOL=True \
  -e TRAINING_DATATOOL=True \
  knucklessg1/data-science-mcp:mcp
```

_Auto-generated from the code-read env surface (`MCP_TOOL_MODE` + package vars) — do not edit._
<!-- MCP-CONFIG-EXAMPLES:END -->

<!-- BEGIN GENERATED: additional-deployment-options -->
### Additional Deployment Options

`data-science-mcp` can also run as a **local container** (Docker / Podman / `uv`) or be
consumed from a **remote deployment**. The
[Deployment guide](https://knuckles-team.github.io/data-science-mcp/deployment/) has full, copy-paste
`mcp_config.json` for all four transports — **stdio**, **streamable-http**,
**local container / uv**, and **remote URL**:

- **Local container / uv** — launch the server from `mcp_config.json` via `uvx`,
  `docker run`, or `podman run`, or point at a local streamable-http container by `url`.
- **Remote URL** — connect to a server deployed behind Caddy at
  `http://data-science-mcp.arpa/mcp` using the `"url"` key.
<!-- END GENERATED: additional-deployment-options -->

---

## Environment Variables

<!-- ENV-VARS-TABLE:START -->

#### Package environment variables

| Variable | Example | Description |
|----------|---------|-------------|
| `HOST` | `0.0.0.0` |  |
| `PORT` | `8000` |  |
| `TRANSPORT` | `stdio` | options: stdio, streamable-http, sse |
| `ENABLE_OTEL` | `True` |  |
| `OTEL_EXPORTER_OTLP_ENDPOINT` | `http://localhost:8080/api/public/otel` |  |
| `OTEL_EXPORTER_OTLP_PUBLIC_KEY` | `pk-...` |  |
| `OTEL_EXPORTER_OTLP_SECRET_KEY` | `sk-...` |  |
| `OTEL_EXPORTER_OTLP_PROTOCOL` | `http/protobuf` |  |
| `EUNOMIA_TYPE` | `none` | options: none, embedded, remote |
| `EUNOMIA_POLICY_FILE` | `mcp_policies.json` |  |
| `EUNOMIA_REMOTE_URL` | `http://eunomia-server:8000` |  |
| `DATA_SCIENCE_MCP_URL` | `http://localhost:8080` |  |
| `DATA_SCIENCE_MCP_TOKEN` | `your_token_here` |  |
| `DATA_SCIENCE_MCP_SSL_VERIFY` | — | TLS verification for the upstream client. Set to a CA bundle path or False to disable. |
| `DATA_SCIENCE_MCP_VERIFY` | `True` | Legacy alias for DATA_SCIENCE_MCP_SSL_VERIFY (truthy enables verification). |
| `INFERENCE_BACKEND` | `vllm` | options: vllm, sglang |
| `INFERENCE_BASE_URL` | — | Base URL of the running inference server, e.g. http://host:30000 |
| `INFERENCE_MODEL` | — | Served model id exposed by the inference server. |
| `INFERENCE_API_KEY` | `EMPTY` | Bearer token for the inference server (default "EMPTY" for local servers). |
| `EPISTEMIC_GRAPH_SOCKET` | — | Unix domain socket path to the epistemic-graph engine. |
| `GRAPH_SERVICE_SOCKET` | — | Alternate UDS env var honored by the engine client. |
| `EPISTEMIC_GRAPH_TCP` | — | TCP host:port for the epistemic-graph engine (takes precedence over the socket). |
| `DSM_NEAR_PAIRS_LOCAL_MAX` | `20000` | Cap on local O(n^2) near-pair fallback before requiring the Rust path (0 disables the cap). |
| `MODEL_TRAININGTOOL` | `True` |  |
| `MODEL_EVOLUTIONTOOL` | `True` |  |
| `INTERPRETABILITYTOOL` | `True` |  |
| `DATA_MANAGEMENTTOOL` | `True` |  |
| `DATA_ENGINETOOL` | `True` |  |
| `QUANTTOOL` | `True` |  |
| `TRAINERTOOL` | `True` | Sub-surfaces of model-training; the code gates these via MODEL_TRAININGTOOL. |
| `TRAINING_DATATOOL` | `True` |  |
| `KERNEL_SPECIALIZETOOL` | `True` |  |

#### Inherited agent-utilities variables (apply to every connector)

| Variable | Example | Description |
|----------|---------|-------------|
| `MCP_TOOL_MODE` | `condensed` | Tool surface: `condensed` | `verbose` | `both` |
| `MCP_ENABLED_TOOLS` | — | Comma-separated tool allow-list |
| `MCP_DISABLED_TOOLS` | — | Comma-separated tool deny-list |
| `MCP_ENABLED_TAGS` | — | Comma-separated tag allow-list |
| `MCP_DISABLED_TAGS` | — | Comma-separated tag deny-list |
| `MCP_CLIENT_AUTH` | — | Outbound MCP auth (`oidc-client-credentials` for fleet calls) |
| `OIDC_CLIENT_ID` | — | OIDC client id (service-account auth) |
| `OIDC_CLIENT_SECRET` | — | OIDC client secret (service-account auth) |
| `DEBUG` | `False` | Verbose logging |
| `PYTHONUNBUFFERED` | `1` | Unbuffered stdout (recommended in containers) |
| `MCP_URL` | `http://localhost:8000/mcp` | URL of the MCP server the agent connects to |
| `PROVIDER` | `openai` | LLM provider for the agent |
| `MODEL_ID` | `gpt-4o` | Model id for the agent |
| `ENABLE_WEB_UI` | `True` | Serve the AG-UI web interface |

_32 package + 14 inherited variable(s). Auto-generated from `.env.example` + the shared agent-utilities set — do not edit._
<!-- ENV-VARS-TABLE:END -->


Every variable the server reads, grouped by purpose.

### MCP server / transport
| Variable | Description | Default |
|----------|-------------|---------|
| `TRANSPORT` | `stdio`, `streamable-http`, or `sse` | `stdio` |
| `HOST` | Bind host (HTTP transports) | `0.0.0.0` |
| `PORT` | Bind port (HTTP transports) | `8000` |
| `MCP_TOOL_MODE` | Tool surface: `condensed`, `verbose`, or `both` | `condensed` |
| `MCP_ENABLED_TOOLS` / `MCP_DISABLED_TOOLS` | Comma-separated tool allow/deny list | — |
| `MCP_ENABLED_TAGS` / `MCP_DISABLED_TAGS` | Comma-separated tag allow/deny list | — |
| `DEBUG` | Verbose logging | `False` |
| `PYTHONUNBUFFERED` | Unbuffered stdout (recommended in containers) | `1` |

### Connection
| Variable | Description | Default |
|----------|-------------|---------|
| `DATA_SCIENCE_MCP_URL` | Base service URL | `http://localhost:8080` |
| `DATA_SCIENCE_MCP_TOKEN` | API token | — |

### Training / inference backend (full `[training]` extra)
| Variable | Description | Default |
|----------|-------------|---------|
| `INFERENCE_BACKEND` | Served-model rollout backend (`vllm` or `sglang`) | — |
| `INFERENCE_BASE_URL` | OpenAI-compatible inference server base URL | — |

### Tool toggles
Each action-routed tool can be disabled individually via its toggle env var (set to `false`).
The full list is in the [Available MCP Tools](#available-mcp-tools) table above.
| Variable | Tools |
|----------|-------|
| `MODEL_TRAININGTOOL` | training / fit / eval / corpus / kernel tools |
| `MODEL_EVOLUTIONTOOL` | Pareto-frontier evolution + model ranking |
| `INTERPRETABILITYTOOL` | interpretability test generation + grading suite |
| `DATA_MANAGEMENTTOOL` | dataset load / describe / split |
| `DATA_ENGINETOOL` | corpus curation / dedup / decontaminate / lineage |
| `QUANTTOOL` | quant compute kernels |

### Telemetry & governance
| Variable | Description | Default |
|----------|-------------|---------|
| `ENABLE_OTEL` | Enable OpenTelemetry export | `True` |
| `OTEL_EXPORTER_OTLP_ENDPOINT` | OTLP collector endpoint | — |
| `OTEL_EXPORTER_OTLP_PUBLIC_KEY` / `OTEL_EXPORTER_OTLP_SECRET_KEY` | OTLP auth keys | — |
| `OTEL_EXPORTER_OTLP_PROTOCOL` | OTLP protocol (e.g. `http/protobuf`) | — |
| `EUNOMIA_TYPE` | Authorization mode: `none`, `embedded`, `remote` | `none` |
| `EUNOMIA_POLICY_FILE` | Embedded policy file | `mcp_policies.json` |
| `EUNOMIA_REMOTE_URL` | Remote Eunomia server URL | — |

### Agent CLI (full `[agent]` runtime only)
| Variable | Description | Default |
|----------|-------------|---------|
| `MCP_URL` | URL of the MCP server the agent connects to | `http://localhost:8000/mcp` |
| `PROVIDER` | LLM provider (e.g. `openai`) | `openai` |
| `MODEL_ID` | Model id (e.g. `gpt-4o`) | `gpt-4o` |
| `ENABLE_WEB_UI` | Serve the AG-UI web interface | `True` |

See [`.env.example`](.env.example) for a copy-paste starting point.

## Agent

This repository features a fully integrated Pydantic AI Graph Agent. It communicates over the **Agent Control Protocol (ACP)** and interacts seamlessly with the **Agent Web UI (AG-UI)** and Terminal interface.

### Running the Agent CLI
To start the interactive command-line agent:

```bash
# Set credentials
export DATA_SCIENCE_MCP_URL="your_value"
export DATA_SCIENCE_MCP_TOKEN="your_value"

# Run the agent server
data-science-agent --provider openai --model-id gpt-4o
```

### Docker Compose Orchestration
The following `docker/agent.compose.yml` configures the Agent, Web UI, and Terminal Interface together:

```yaml
version: '3.8'

services:
  data-science-mcp-mcp:
    image: knucklessg1/data-science-mcp:mcp
    container_name: data-science-mcp-mcp
    hostname: data-science-mcp-mcp
    restart: always
    env_file:
      - ../.env
    environment:
      - PYTHONUNBUFFERED=1
      - HOST=0.0.0.0
      - PORT=8000
      - TRANSPORT=streamable-http
    ports:
      - "8000:8000"
    healthcheck:
      test: ["CMD", "python3", "-c", "import urllib.request; urllib.request.urlopen('http://localhost:8000/health')"]
      interval: 30s
      timeout: 10s
      retries: 3
      start_period: 10s
    logging:
      driver: json-file
      options:
        max-size: "10m"
        max-file: "3"

  data-science-mcp-agent:
    image: knucklessg1/data-science-mcp:latest
    container_name: data-science-mcp-agent
    hostname: data-science-mcp-agent
    restart: always
    depends_on:
      - data-science-mcp-mcp
    env_file:
      - ../.env
    command: [ "data-science-agent" ]
    environment:
      - PYTHONUNBUFFERED=1
      - HOST=0.0.0.0
      - PORT=9004
      - MCP_URL=http://data-science-mcp-mcp:8000/mcp
      - PROVIDER=${PROVIDER:-openai}
      - MODEL_ID=${MODEL_ID:-gpt-4o}
      - ENABLE_WEB_UI=True
      - ENABLE_OTEL=True
    ports:
      - "9004:9004"
    healthcheck:
      test: ["CMD", "python3", "-c", "import urllib.request; urllib.request.urlopen('http://localhost:9004/health')"]
      interval: 30s
      timeout: 10s
      retries: 3
      start_period: 10s
    logging:
      driver: json-file
      options:
        max-size: "10m"
        max-file: "3"

```

Detailed graph node architecture explanations, custom skill configurations, and agentic trace guides are available in [docs/agent.md](docs/agent.md).

---

## Security & Governance

Built directly upon the enterprise-ready [`agent-utilities`](https://github.com/Knuckles-Team/agent-utilities) core, standard security parameters are fully supported:

### Access Control & Policy Enforcement
- **Eunomia Policies:** Fine-grained, policy-driven tool authorization. Supports `none`, local `embedded` (`mcp_policies.json`), or centralized `remote` modes.
- **OIDC Token Delegation:** Compliant with RFC 8693 token exchange for flowing authenticating user credentials from Web UI / ACP → Agent → MCP.
- **Scoped Credentials:** Execution context runs restricted to the specific caller identity.

### Runtime Security Grid
| Feature | Functionality | Enablement |
|---------|---------------|------------|
| **Tool Guard** | Sensitivity inspection with human-in-the-loop validation | Enabled by default |
| **Prompt Injection Defense** | Input scanning, repetition monitoring, and recursive loop blocks | Enabled by default |
| **Context Safety Guard** | Stuck-loop detectors and contextual overflow preemptive alerts | Enabled by default |

---

## Installation

Pick the extra that matches what you want to run:

| Extra | Installs | Use when |
|-------|----------|----------|
| `data-science-mcp[mcp]` | Slim MCP server only (`agent-utilities[mcp]` — FastMCP/FastAPI) + the core epistemic-graph engine | You only run the **MCP server** (smallest install / image) |
| `data-science-mcp[agent]` | Full agent runtime (`agent-utilities[agent,logfire]` — Pydantic AI) | You run the **integrated agent** |
| `data-science-mcp[all]` | Everything (`mcp` + `agent` + `scikit-learn` sample-dataset loaders) | Development / both surfaces |

Heavy ML extras are opt-in and imported lazily — add them only when needed:
`[training]` (torch/PEFT gradient trainers), `[training-scale]` (DeepSpeed/FlashAttention,
GPU-host only), `[training-fast]` (Liger Triton kernels), `[datasets]` (scikit-learn sample
loaders), `[eval]` (lm-eval), `[tracking]` (MLflow). See **[docs/training.md](docs/training.md)**.

```bash
# MCP server only (recommended for tool hosting — slim deps)
uv pip install "data-science-mcp[mcp]"

# Full agent runtime (Pydantic AI)
uv pip install "data-science-mcp[agent]"

# Everything (development)
uv pip install "data-science-mcp[all]"      # or: python -m pip install "data-science-mcp[all]"
```

### Container images (`:mcp` vs `:agent`)

One multi-stage `docker/Dockerfile` builds two right-sized images, selected by `--target`:

| Image tag | Build target | Contents | Entrypoint |
|-----------|--------------|----------|------------|
| `knucklessg1/data-science-mcp:mcp` | `--target mcp` | `data-science-mcp[mcp]` — **slim**, no `pydantic-ai`/`dspy`/`llama-index`/`tree-sitter` | `data-science-mcp` |
| `knucklessg1/data-science-mcp:latest` | `--target agent` (default) | `data-science-mcp[agent]` — **full** agent runtime | `data-science-agent` |

```bash
docker build --target mcp   -t knucklessg1/data-science-mcp:mcp    docker/   # slim MCP server
docker build --target agent -t knucklessg1/data-science-mcp:latest docker/   # full agent
```

`docker/mcp.compose.yml` runs the slim `:mcp` server; `docker/agent.compose.yml` runs the
agent (`:latest`) with a co-located `:mcp` sidecar.

### Knowledge-graph database (`epistemic-graph`)

`data-science-mcp` depends directly on the **epistemic-graph** compute engine
(`epistemic-graph[datascience]`) — its model-training / evaluation / evolution compute runs in
that Rust engine, so the engine is a core dependency in **every** extra (including `[mcp]`). For
production — or to share one knowledge graph across multiple agents — run **epistemic-graph as
its own database container** and point the server at it instead of embedding it. Deployment
recipes (single-node + Raft HA), connection config, and the full database architecture (with
diagrams) are documented in the
[epistemic-graph deployment guide](https://knuckles-team.github.io/epistemic-graph/deployment/).

---

## Documentation

The complete documentation is published as the
[official documentation site](https://knuckles-team.github.io/data-science-mcp/) and
is the recommended reference for installation, deployment, and day-to-day operation.

| Page | Contents |
|---|---|
| [Installation](https://knuckles-team.github.io/data-science-mcp/installation/) | pip, source, extras, prebuilt Docker image |
| [Deployment](https://knuckles-team.github.io/data-science-mcp/deployment/) | run the MCP server and A2A agent, Compose, Caddy + Technitium, env config |
| [Usage](https://knuckles-team.github.io/data-science-mcp/usage/) | the MCP tools, the `MLEngine` Python API, the console scripts |
| [Overview](https://knuckles-team.github.io/data-science-mcp/overview/) | ecosystem role, enterprise readiness, concept registry |
| [Model Training](https://knuckles-team.github.io/data-science-mcp/training/) | SFT/DPO/GRPO corpus, reward engine, gradient trainers |
| [Concepts](https://knuckles-team.github.io/data-science-mcp/concepts/) | concept registry (`CONCEPT:DSCI-*`) |

---

## Repository Owners

<img width="100%" height="180em" src="https://github-readme-stats.vercel.app/api?username=Knucklessg1&show_icons=true&hide_border=true&&count_private=true&include_all_commits=true" />

![GitHub followers](https://img.shields.io/github/followers/Knucklessg1)
![GitHub User's stars](https://img.shields.io/github/stars/Knucklessg1)

---

## Contribute

Contributions are welcome! Please ensure code quality by executing local checks before submitting pull requests:
- Format code using `ruff format .`
- Lint code using `ruff check .`
- Validate type-safety with `mypy .`
- Execute test suites using `pytest`


<!-- BEGIN agent-os-genesis-deploy (generated; do not edit between markers) -->

## Deploy with `agent-os-genesis`

This package can be provisioned for you — skill-guided — by the **`agent-os-genesis`**
universal skill (its *single-package deploy mode*): it picks your install method, seeds
secrets to OpenBao/Vault (or `.env`), trusts your enterprise CA, registers the MCP
server, and verifies it — the same machinery that stands up the whole Agent OS, narrowed
to just this package. Ask your agent to **"deploy `data-science-mcp` with agent-os-genesis"**.

| Install mode | Command |
|------|---------|
| Bare-metal, prod (PyPI) | `uvx data-science-mcp` · or `uv tool install data-science-mcp` |
| Bare-metal, dev (editable) | `uv pip install -e ".[all]"` · or `pip install -e ".[all]"` |
| Container, prod | deploy `knucklessg1/data-science-mcp:latest` via docker-compose / swarm / podman / podman-compose / kubernetes |
| Container, dev (editable) | deploy `docker/compose.dev.yml` (source-mounted at `/src`; edits live on restart) |

Secrets are read-existing + seeded via `vault_sync` — you are only prompted for what's missing.

<!-- END agent-os-genesis-deploy -->
