data-science-mcp 0.2.0__tar.gz

This diff represents the content of publicly available package versions that have been released to one of the supported registries. The information contained in this diff is provided for informational purposes only and reflects changes between package versions as they appear in their respective public registries.
Files changed (28) hide show
  1. data_science_mcp-0.2.0/LICENSE +21 -0
  2. data_science_mcp-0.2.0/MANIFEST.in +4 -0
  3. data_science_mcp-0.2.0/PKG-INFO +164 -0
  4. data_science_mcp-0.2.0/README.md +140 -0
  5. data_science_mcp-0.2.0/data_science_mcp/__init__.py +70 -0
  6. data_science_mcp-0.2.0/data_science_mcp/__main__.py +6 -0
  7. data_science_mcp-0.2.0/data_science_mcp/agent/AGENTS.md +12 -0
  8. data_science_mcp-0.2.0/data_science_mcp/agent/CRON.md +12 -0
  9. data_science_mcp-0.2.0/data_science_mcp/agent/CRON_LOG.md +5 -0
  10. data_science_mcp-0.2.0/data_science_mcp/agent/HEARTBEAT.md +28 -0
  11. data_science_mcp-0.2.0/data_science_mcp/agent/IDENTITY.md +15 -0
  12. data_science_mcp-0.2.0/data_science_mcp/agent/MCP_AGENTS.md +13 -0
  13. data_science_mcp-0.2.0/data_science_mcp/agent/USER.md +7 -0
  14. data_science_mcp-0.2.0/data_science_mcp/agent/icon.png +0 -0
  15. data_science_mcp-0.2.0/data_science_mcp/agent/mcp_config.json +11 -0
  16. data_science_mcp-0.2.0/data_science_mcp/agent_server.py +76 -0
  17. data_science_mcp-0.2.0/data_science_mcp/auth.py +47 -0
  18. data_science_mcp-0.2.0/data_science_mcp/mcp_server.py +719 -0
  19. data_science_mcp-0.2.0/data_science_mcp/ml_engine.py +468 -0
  20. data_science_mcp-0.2.0/data_science_mcp.egg-info/PKG-INFO +164 -0
  21. data_science_mcp-0.2.0/data_science_mcp.egg-info/SOURCES.txt +26 -0
  22. data_science_mcp-0.2.0/data_science_mcp.egg-info/dependency_links.txt +1 -0
  23. data_science_mcp-0.2.0/data_science_mcp.egg-info/entry_points.txt +3 -0
  24. data_science_mcp-0.2.0/data_science_mcp.egg-info/requires.txt +10 -0
  25. data_science_mcp-0.2.0/data_science_mcp.egg-info/top_level.txt +2 -0
  26. data_science_mcp-0.2.0/pyproject.toml +45 -0
  27. data_science_mcp-0.2.0/requirements.txt +5 -0
  28. data_science_mcp-0.2.0/setup.cfg +4 -0
@@ -0,0 +1,21 @@
1
+ MIT License
2
+
3
+ Copyright (c) 2026 Audel Rouhi
4
+
5
+ Permission is hereby granted, free of charge, to any person obtaining a copy
6
+ of this software and associated documentation files (the "Software"), to deal
7
+ in the Software without restriction, including without limitation the rights
8
+ to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
9
+ copies of the Software, and to permit persons to whom the Software is
10
+ furnished to do so, subject to the following conditions:
11
+
12
+ The above copyright notice and this permission notice shall be included in all
13
+ copies or substantial portions of the Software.
14
+
15
+ THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
16
+ IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
17
+ FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
18
+ AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
19
+ LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
20
+ OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
21
+ SOFTWARE.
@@ -0,0 +1,4 @@
1
+ include LICENSE
2
+ include README.md
3
+ include requirements.txt
4
+ recursive-include data_science_mcp *.md *.json *.yaml *.yml *.py
@@ -0,0 +1,164 @@
1
+ Metadata-Version: 2.4
2
+ Name: data-science-mcp
3
+ Version: 0.2.0
4
+ Summary: Data Science MCP Server — Model training, evaluation, and evolution tools for agentic ML workflows. Integrates with agent-utilities IModelEvolver (CONCEPT:AHE-3.15).
5
+ Author-email: Audel Rouhi <knucklessg1@gmail.com>
6
+ License: MIT
7
+ Classifier: Development Status :: 4 - Beta
8
+ Classifier: License :: OSI Approved :: MIT License
9
+ Classifier: Environment :: Console
10
+ Classifier: Operating System :: POSIX :: Linux
11
+ Classifier: Programming Language :: Python :: 3
12
+ Requires-Python: >=3.10
13
+ Description-Content-Type: text/markdown
14
+ License-File: LICENSE
15
+ Requires-Dist: agent-utilities[mcp]>=0.2.12
16
+ Requires-Dist: scikit-learn>=1.5.0
17
+ Requires-Dist: pandas>=2.0.0
18
+ Requires-Dist: numpy>=1.24.0
19
+ Provides-Extra: agent
20
+ Requires-Dist: agent-utilities[agent,logfire]>=0.2.12; extra == "agent"
21
+ Provides-Extra: all
22
+ Requires-Dist: agent-utilities[agent,logfire,mcp]>=0.2.12; extra == "all"
23
+ Dynamic: license-file
24
+
25
+ # Data Science MCP - A2A | AG-UI | MCP
26
+
27
+ ![PyPI - Version](https://img.shields.io/pypi/v/data-science-mcp)
28
+ ![MCP Server](https://badge.mcpx.dev?type=server 'MCP Server')
29
+ ![PyPI - Downloads](https://img.shields.io/pypi/dd/data-science-mcp)
30
+ ![GitHub Repo stars](https://img.shields.io/github/stars/Knuckles-Team/data-science-mcp)
31
+ ![GitHub forks](https://img.shields.io/github/forks/Knuckles-Team/data-science-mcp)
32
+ ![GitHub contributors](https://img.shields.io/github/contributors/Knuckles-Team/data-science-mcp)
33
+ ![PyPI - License](https://img.shields.io/pypi/l/data-science-mcp)
34
+ ![GitHub](https://img.shields.io/github/license/Knuckles-Team/data-science-mcp)
35
+
36
+ ![GitHub last commit (by committer)](https://img.shields.io/github/last-commit/Knuckles-Team/data-science-mcp)
37
+ ![GitHub pull requests](https://img.shields.io/github/issues-pr/Knuckles-Team/data-science-mcp)
38
+ ![GitHub closed pull requests](https://img.shields.io/github/issues-pr-closed/Knuckles-Team/data-science-mcp)
39
+ ![GitHub issues](https://img.shields.io/github/issues/Knuckles-Team/data-science-mcp)
40
+
41
+ ![GitHub top language](https://img.shields.io/github/languages/top/Knuckles-Team/data-science-mcp)
42
+ ![GitHub language count](https://img.shields.io/github/languages/count/Knuckles-Team/data-science-mcp)
43
+ ![GitHub repo size](https://img.shields.io/github/repo-size/Knuckles-Team/data-science-mcp)
44
+ ![GitHub repo file count (file type)](https://img.shields.io/github/directory-file-count/Knuckles-Team/data-science-mcp)
45
+ ![PyPI - Wheel](https://img.shields.io/pypi/wheel/data-science-mcp)
46
+ ![PyPI - Implementation](https://img.shields.io/pypi/implementation/data-science-mcp)
47
+
48
+ *Version: 0.2.0*
49
+
50
+ ## Overview
51
+
52
+ **Data Science MCP MCP Server + A2A Agent**
53
+
54
+ Data Science MCP Server — Model training, evaluation, and evolution tools for agentic ML workflows. Integrates with agent-utilities IModelEvolver (CONCEPT:AHE-3.15).
55
+
56
+ This repository is actively maintained - Contributions are welcome!
57
+
58
+ ## MCP
59
+
60
+ ### Using as an MCP Server
61
+
62
+ The MCP Server can be run in two modes: `stdio` (for local testing) or `http` (for networked access).
63
+
64
+ #### Environment Variables
65
+
66
+ * `DATA_SCIENCE_MCP_URL`: The URL of the target service.
67
+ * `DATA_SCIENCE_MCP_TOKEN`: The API token or access token.
68
+
69
+ #### Run in stdio mode (default):
70
+ ```bash
71
+ export DATA_SCIENCE_MCP_URL="http://localhost:8080"
72
+ export DATA_SCIENCE_MCP_TOKEN="your_token"
73
+ data-science-mcp --transport "stdio"
74
+ ```
75
+
76
+ #### Run in HTTP mode:
77
+ ```bash
78
+ export DATA_SCIENCE_MCP_URL="http://localhost:8080"
79
+ export DATA_SCIENCE_MCP_TOKEN="your_token"
80
+ data-science-mcp --transport "http" --host "0.0.0.0" --port "8000"
81
+ ```
82
+
83
+ ## A2A Agent
84
+
85
+ ### Run A2A Server
86
+ ```bash
87
+ export DATA_SCIENCE_MCP_URL="http://localhost:8080"
88
+ export DATA_SCIENCE_MCP_TOKEN="your_token"
89
+ data-science-agent --provider openai --model-id gpt-4o --api-key sk-...
90
+ ```
91
+
92
+ ## Docker
93
+
94
+ ### Build
95
+
96
+ ```bash
97
+ docker build -t data-science-mcp .
98
+ ```
99
+
100
+ ### Run MCP Server
101
+
102
+ ```bash
103
+ docker run -d \
104
+ --name data-science-mcp \
105
+ -p 8000:8000 \
106
+ -e TRANSPORT=http \
107
+ -e DATA_SCIENCE_MCP_URL="http://your-service:8080" \
108
+ -e DATA_SCIENCE_MCP_TOKEN="your_token" \
109
+ knucklessg1/data-science-mcp:latest
110
+ ```
111
+
112
+ ### Deploy with Docker Compose
113
+
114
+ ```yaml
115
+ services:
116
+ data-science-mcp:
117
+ image: knucklessg1/data-science-mcp:latest
118
+ environment:
119
+ - HOST=0.0.0.0
120
+ - PORT=8000
121
+ - TRANSPORT=http
122
+ - DATA_SCIENCE_MCP_URL=http://your-service:8080
123
+ - DATA_SCIENCE_MCP_TOKEN=your_token
124
+ ports:
125
+ - 8000:8000
126
+ ```
127
+
128
+ #### Configure `mcp.json` for AI Integration (e.g. Claude Desktop)
129
+
130
+ ```json
131
+ {
132
+ "mcpServers": {
133
+ "data-science": {
134
+ "command": "uv",
135
+ "args": [
136
+ "run",
137
+ "--with",
138
+ "data-science-mcp",
139
+ "data-science-mcp"
140
+ ],
141
+ "env": {
142
+ "DATA_SCIENCE_MCP_URL": "http://your-service:8080",
143
+ "DATA_SCIENCE_MCP_TOKEN": "your_token"
144
+ }
145
+ }
146
+ }
147
+ }
148
+ ```
149
+
150
+ ## Install Python Package
151
+
152
+ ```bash
153
+ python -m pip install data-science-mcp
154
+ ```
155
+ ```bash
156
+ uv pip install data-science-mcp
157
+ ```
158
+
159
+ ## Repository Owners
160
+
161
+ <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" />
162
+
163
+ ![GitHub followers](https://img.shields.io/github/followers/Knucklessg1)
164
+ ![GitHub User's stars](https://img.shields.io/github/stars/Knucklessg1)
@@ -0,0 +1,140 @@
1
+ # Data Science MCP - A2A | AG-UI | MCP
2
+
3
+ ![PyPI - Version](https://img.shields.io/pypi/v/data-science-mcp)
4
+ ![MCP Server](https://badge.mcpx.dev?type=server 'MCP Server')
5
+ ![PyPI - Downloads](https://img.shields.io/pypi/dd/data-science-mcp)
6
+ ![GitHub Repo stars](https://img.shields.io/github/stars/Knuckles-Team/data-science-mcp)
7
+ ![GitHub forks](https://img.shields.io/github/forks/Knuckles-Team/data-science-mcp)
8
+ ![GitHub contributors](https://img.shields.io/github/contributors/Knuckles-Team/data-science-mcp)
9
+ ![PyPI - License](https://img.shields.io/pypi/l/data-science-mcp)
10
+ ![GitHub](https://img.shields.io/github/license/Knuckles-Team/data-science-mcp)
11
+
12
+ ![GitHub last commit (by committer)](https://img.shields.io/github/last-commit/Knuckles-Team/data-science-mcp)
13
+ ![GitHub pull requests](https://img.shields.io/github/issues-pr/Knuckles-Team/data-science-mcp)
14
+ ![GitHub closed pull requests](https://img.shields.io/github/issues-pr-closed/Knuckles-Team/data-science-mcp)
15
+ ![GitHub issues](https://img.shields.io/github/issues/Knuckles-Team/data-science-mcp)
16
+
17
+ ![GitHub top language](https://img.shields.io/github/languages/top/Knuckles-Team/data-science-mcp)
18
+ ![GitHub language count](https://img.shields.io/github/languages/count/Knuckles-Team/data-science-mcp)
19
+ ![GitHub repo size](https://img.shields.io/github/repo-size/Knuckles-Team/data-science-mcp)
20
+ ![GitHub repo file count (file type)](https://img.shields.io/github/directory-file-count/Knuckles-Team/data-science-mcp)
21
+ ![PyPI - Wheel](https://img.shields.io/pypi/wheel/data-science-mcp)
22
+ ![PyPI - Implementation](https://img.shields.io/pypi/implementation/data-science-mcp)
23
+
24
+ *Version: 0.2.0*
25
+
26
+ ## Overview
27
+
28
+ **Data Science MCP MCP Server + A2A Agent**
29
+
30
+ Data Science MCP Server — Model training, evaluation, and evolution tools for agentic ML workflows. Integrates with agent-utilities IModelEvolver (CONCEPT:AHE-3.15).
31
+
32
+ This repository is actively maintained - Contributions are welcome!
33
+
34
+ ## MCP
35
+
36
+ ### Using as an MCP Server
37
+
38
+ The MCP Server can be run in two modes: `stdio` (for local testing) or `http` (for networked access).
39
+
40
+ #### Environment Variables
41
+
42
+ * `DATA_SCIENCE_MCP_URL`: The URL of the target service.
43
+ * `DATA_SCIENCE_MCP_TOKEN`: The API token or access token.
44
+
45
+ #### Run in stdio mode (default):
46
+ ```bash
47
+ export DATA_SCIENCE_MCP_URL="http://localhost:8080"
48
+ export DATA_SCIENCE_MCP_TOKEN="your_token"
49
+ data-science-mcp --transport "stdio"
50
+ ```
51
+
52
+ #### Run in HTTP mode:
53
+ ```bash
54
+ export DATA_SCIENCE_MCP_URL="http://localhost:8080"
55
+ export DATA_SCIENCE_MCP_TOKEN="your_token"
56
+ data-science-mcp --transport "http" --host "0.0.0.0" --port "8000"
57
+ ```
58
+
59
+ ## A2A Agent
60
+
61
+ ### Run A2A Server
62
+ ```bash
63
+ export DATA_SCIENCE_MCP_URL="http://localhost:8080"
64
+ export DATA_SCIENCE_MCP_TOKEN="your_token"
65
+ data-science-agent --provider openai --model-id gpt-4o --api-key sk-...
66
+ ```
67
+
68
+ ## Docker
69
+
70
+ ### Build
71
+
72
+ ```bash
73
+ docker build -t data-science-mcp .
74
+ ```
75
+
76
+ ### Run MCP Server
77
+
78
+ ```bash
79
+ docker run -d \
80
+ --name data-science-mcp \
81
+ -p 8000:8000 \
82
+ -e TRANSPORT=http \
83
+ -e DATA_SCIENCE_MCP_URL="http://your-service:8080" \
84
+ -e DATA_SCIENCE_MCP_TOKEN="your_token" \
85
+ knucklessg1/data-science-mcp:latest
86
+ ```
87
+
88
+ ### Deploy with Docker Compose
89
+
90
+ ```yaml
91
+ services:
92
+ data-science-mcp:
93
+ image: knucklessg1/data-science-mcp:latest
94
+ environment:
95
+ - HOST=0.0.0.0
96
+ - PORT=8000
97
+ - TRANSPORT=http
98
+ - DATA_SCIENCE_MCP_URL=http://your-service:8080
99
+ - DATA_SCIENCE_MCP_TOKEN=your_token
100
+ ports:
101
+ - 8000:8000
102
+ ```
103
+
104
+ #### Configure `mcp.json` for AI Integration (e.g. Claude Desktop)
105
+
106
+ ```json
107
+ {
108
+ "mcpServers": {
109
+ "data-science": {
110
+ "command": "uv",
111
+ "args": [
112
+ "run",
113
+ "--with",
114
+ "data-science-mcp",
115
+ "data-science-mcp"
116
+ ],
117
+ "env": {
118
+ "DATA_SCIENCE_MCP_URL": "http://your-service:8080",
119
+ "DATA_SCIENCE_MCP_TOKEN": "your_token"
120
+ }
121
+ }
122
+ }
123
+ }
124
+ ```
125
+
126
+ ## Install Python Package
127
+
128
+ ```bash
129
+ python -m pip install data-science-mcp
130
+ ```
131
+ ```bash
132
+ uv pip install data-science-mcp
133
+ ```
134
+
135
+ ## Repository Owners
136
+
137
+ <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" />
138
+
139
+ ![GitHub followers](https://img.shields.io/github/followers/Knucklessg1)
140
+ ![GitHub User's stars](https://img.shields.io/github/stars/Knucklessg1)
@@ -0,0 +1,70 @@
1
+ #!/usr/bin/env python
2
+ # coding: utf-8
3
+
4
+ import importlib
5
+ import inspect
6
+ import warnings
7
+ from typing import List
8
+
9
+ # Suppress RequestsDependencyWarning due to chardet 6.x / requests 2.32.x mismatch
10
+ # Centralized here to ensure it runs before any sub-package imports
11
+ warnings.filterwarnings("ignore", message=".*urllib3.*or chardet.*")
12
+
13
+ __all__: List[str] = []
14
+
15
+ CORE_MODULES = [
16
+ "data_science_mcp.api_client",
17
+ ]
18
+
19
+ OPTIONAL_MODULES = {
20
+ "data_science_mcp.agent_server": "agent",
21
+ "data_science_mcp.mcp_server": "mcp",
22
+ }
23
+
24
+
25
+ def _import_module_safely(module_name: str):
26
+ """Try to import a module and return it, or None if not available."""
27
+ try:
28
+ return importlib.import_module(module_name)
29
+ except ImportError:
30
+ return None
31
+
32
+
33
+ def _expose_members(module):
34
+ """Expose public classes and functions from a module into globals and __all__."""
35
+ for name, obj in inspect.getmembers(module):
36
+ if (inspect.isclass(obj) or inspect.isfunction(obj)) and not name.startswith(
37
+ "_"
38
+ ):
39
+ globals()[name] = obj
40
+ __all__.append(name)
41
+
42
+
43
+ for module_name in CORE_MODULES:
44
+ try:
45
+ module = importlib.import_module(module_name)
46
+ _expose_members(module)
47
+ except ImportError:
48
+ pass
49
+
50
+ for module_name, extra_name in OPTIONAL_MODULES.items():
51
+ module = _import_module_safely(module_name)
52
+ if module is not None:
53
+ _expose_members(module)
54
+ globals()[f"_{extra_name.upper()}_AVAILABLE"] = True
55
+ else:
56
+ globals()[f"_{extra_name.upper()}_AVAILABLE"] = False
57
+
58
+ _MCP_AVAILABLE = OPTIONAL_MODULES.get("data_science_mcp.mcp_server") in [
59
+ m.__name__ for m in globals().values() if hasattr(m, "__name__")
60
+ ]
61
+ _AGENT_AVAILABLE = "data_science_mcp.agent_server" in globals()
62
+
63
+ __all__.extend(["_MCP_AVAILABLE", "_AGENT_AVAILABLE"])
64
+
65
+
66
+ """
67
+ data-science-mcp
68
+
69
+ Data Science MCP Server — Model training, evaluation, and evolution tools for agentic ML workflows. Integrates with agent-utilities IModelEvolver (CONCEPT:AHE-3.15).
70
+ """
@@ -0,0 +1,6 @@
1
+ #!/usr/bin/python
2
+ # coding: utf-8
3
+ from data_science_mcp.agent_server import agent_server
4
+
5
+ if __name__ == "__main__":
6
+ agent_server()
@@ -0,0 +1,12 @@
1
+ # AGENTS.md - Known A2A Peer Agents
2
+ Last updated: 2026-05-06
3
+
4
+ This file is the local registry of other A2A agents this agent can discover and call.
5
+
6
+ ## Registered A2A Peers
7
+
8
+ | Name | Endpoint URL | Description | Capabilities | Auth | Notes / Last Connected |
9
+ |-----------------|---------------------------------|--------------------------------------|----------------------------------|-----------|------------------------|
10
+ | SearchMaster | http://search-agent:9000/a2a | Advanced web researcher | web_search, summarize, browse | none | 2026-05-06 |
11
+
12
+ *Add new rows manually or let the agent call `register_a2a_peer(...)`.*
@@ -0,0 +1,12 @@
1
+ # CRON.md - Persistent Scheduled Tasks
2
+ Last updated: 2026-05-06
3
+
4
+ ## Active Tasks
5
+
6
+ | ID | Name | Interval (min) | Prompt | Last run | Next approx |
7
+ |-------------|-------------------|----------------|-------------------------------------|-------------------|-------------|
8
+ | heartbeat | Heartbeat | 30 | @HEARTBEAT.md | — | — |
9
+ | log-cleanup | Log Cleanup | 720 | __internal:cleanup_cron_log | — | — |
10
+
11
+ *Edit this table to add/remove tasks. The agent reloads it periodically.*
12
+ *Use `@filename.md` in the Prompt column to load a multi-line prompt from a workspace file.*
@@ -0,0 +1,5 @@
1
+ # CRON_LOG.md - Scheduled Task History
2
+ Last updated: 2026-05-06
3
+
4
+ | Timestamp | Task ID | Status | Message |
5
+ |-----------|---------|--------|---------|
@@ -0,0 +1,28 @@
1
+ # Heartbeat — Periodic Self-Check
2
+
3
+ You are running a scheduled heartbeat. Perform these checks and report results concisely.
4
+
5
+ ## Checks
6
+
7
+ 1. **Tool Availability** — Call `list_tools` or equivalent to verify your MCP tools are reachable. Report any connection failures.
8
+ 2. **Memory Review** — Query the **Knowledge Graph** for any pending follow-up tasks, architectural decisions, or action items.
9
+ 3. **Cron Log** — Read `CRON_LOG.md` and check for recent errors (❌). Summarize any failures from the last 24 hours.
10
+ 4. **Peer Agents** — Read `AGENTS.md` and note if any registered peers need attention.
11
+ 5. **Domain-Specific Checks**:
12
+ - **Service Health**: Check service health status and scan recent logs for critical errors using available tools.
13
+ 6. **Self-Diagnostics** — Report your current model, available tool count, and any anomalies.
14
+
15
+ ## Response Format
16
+
17
+ If everything is healthy:
18
+ ```
19
+ HEARTBEAT_OK — All systems nominal. [tool_count] tools available. No pending actions.
20
+ ```
21
+
22
+ If issues found:
23
+ ```
24
+ HEARTBEAT_ALERT — [summary of issues found]
25
+ - Issue 1: ...
26
+ - Issue 2: ...
27
+ - Action needed: ...
28
+ ```
@@ -0,0 +1,15 @@
1
+ # IDENTITY.md - Data Science MCP Agent Identity
2
+
3
+ ## [default]
4
+ * **Name:** Data Science MCP Agent
5
+ * **Role:** Data Science MCP Server — Model training, evaluation, and evolution tools for agentic ML workflows. Integrates with agent-utilities IModelEvolver (CONCEPT:AHE-3.15).
6
+ * **Emoji:** 🤖
7
+
8
+ ### System Prompt
9
+ You are the Data Science MCP Agent.
10
+ You must always first run `list_skills` to show all skills.
11
+ Then, use the `mcp-client` universal skill and check the reference documentation for `data-science-mcp.md` to discover the exact tags and tools available for your capabilities.
12
+
13
+ ### Capabilities
14
+ - **MCP Operations**: Leverage the `mcp-client` skill to interact with the target MCP server. Refer to `data-science-mcp.md` for specific tool capabilities.
15
+ - **Custom Agent**: Handle custom tasks or general tasks.
@@ -0,0 +1,13 @@
1
+ # MCP_AGENTS.md - Dynamic Agent Registry
2
+
3
+ This file tracks the generated agents from MCP servers. You can manually modify the 'Tools' list to customize agent expertise.
4
+
5
+ ## Agent Mapping Table
6
+
7
+ | Name | Description | System Prompt | Tools | Tag | Source MCP |
8
+ |------|-------------|---------------|-------|-----|------------|
9
+
10
+ ## Tool Inventory Table
11
+
12
+ | Tool Name | Description | Tag | Source |
13
+ |-----------|-------------|-----|--------|
@@ -0,0 +1,7 @@
1
+ # USER.md - About the Human
2
+
3
+ * **Name:** User
4
+ * **Preferred name:** User
5
+ * **Timezone:** America/Chicago
6
+ * **Location:** Chicago, Illinois
7
+ * **Style:** Technical, concise, no fluff
File without changes
@@ -0,0 +1,11 @@
1
+ {
2
+ "mcpServers": {
3
+ "data-science": {
4
+ "command": "data-science-mcp",
5
+ "env": {
6
+ "DATA_SCIENCE_MCP_URL": "${DATA_SCIENCE_MCP_URL:-http://localhost:8080}",
7
+ "DATA_SCIENCE_MCP_TOKEN": "${DATA_SCIENCE_MCP_TOKEN}"
8
+ }
9
+ }
10
+ }
11
+ }
@@ -0,0 +1,76 @@
1
+ #!/usr/bin/python
2
+ # coding: utf-8
3
+ import os
4
+ import sys
5
+ import logging
6
+ import warnings
7
+
8
+ from agent_utilities import (
9
+ build_system_prompt_from_workspace,
10
+ create_agent_parser,
11
+ create_graph_agent_server,
12
+ initialize_workspace,
13
+ load_identity,
14
+ )
15
+
16
+ __version__ = "0.2.0"
17
+
18
+ logging.basicConfig(
19
+ level=logging.INFO,
20
+ format="%(asctime)s - %(name)s - %(levelname)s - %(message)s",
21
+ handlers=[logging.StreamHandler()],
22
+ )
23
+ logger = logging.getLogger(__name__)
24
+
25
+ # Load identity and system prompt from workspace
26
+ initialize_workspace()
27
+ meta = load_identity()
28
+ DEFAULT_AGENT_NAME = os.getenv("DEFAULT_AGENT_NAME", meta.get("name", "Data Science MCP"))
29
+ DEFAULT_AGENT_DESCRIPTION = os.getenv(
30
+ "AGENT_DESCRIPTION",
31
+ meta.get("description", "Data Science MCP Server — Model training, evaluation, and evolution tools for agentic ML workflows. Integrates with agent-utilities IModelEvolver (CONCEPT:AHE-3.15)."),
32
+ )
33
+ DEFAULT_AGENT_SYSTEM_PROMPT = os.getenv(
34
+ "AGENT_SYSTEM_PROMPT",
35
+ meta.get("content") or build_system_prompt_from_workspace(),
36
+ )
37
+
38
+
39
+ def agent_server():
40
+ warnings.filterwarnings("ignore", message=".*urllib3.*or chardet.*")
41
+ warnings.filterwarnings("ignore", category=DeprecationWarning, module="fastmcp")
42
+
43
+ print(f"{DEFAULT_AGENT_NAME} v{__version__}", file=sys.stderr)
44
+ parser = create_agent_parser()
45
+ args = parser.parse_args()
46
+
47
+ if args.debug:
48
+ logging.getLogger().setLevel(logging.DEBUG)
49
+ logger.debug("Debug mode enabled")
50
+
51
+ # Start server using the auto-discovery pattern (from mcp_config.json)
52
+ create_graph_agent_server(
53
+ mcp_url=args.mcp_url,
54
+ mcp_config=args.mcp_config or "mcp_config.json",
55
+ host=args.host,
56
+ port=args.port,
57
+ provider=args.provider,
58
+ model_id=args.model_id,
59
+ router_model=args.model_id,
60
+ agent_model=args.model_id,
61
+ base_url=args.base_url,
62
+ api_key=args.api_key,
63
+ custom_skills_directory=args.custom_skills_directory,
64
+ enable_web_ui=args.web,
65
+ enable_otel=args.otel,
66
+ otel_endpoint=args.otel_endpoint,
67
+ otel_headers=args.otel_headers,
68
+ otel_public_key=args.otel_public_key,
69
+ otel_secret_key=args.otel_secret_key,
70
+ otel_protocol=args.otel_protocol,
71
+ debug=args.debug,
72
+ )
73
+
74
+
75
+ if __name__ == "__main__":
76
+ agent_server()
@@ -0,0 +1,47 @@
1
+ #!/usr/bin/python
2
+ # coding: utf-8
3
+
4
+ import os
5
+ import requests
6
+ import urllib3
7
+
8
+ urllib3.disable_warnings(urllib3.exceptions.InsecureRequestWarning)
9
+
10
+ from agent_utilities.core.exceptions import AuthError, UnauthorizedError
11
+
12
+ # TODO: Import your API wrapper class here
13
+ # from data_science_mcp.api_client import DataScienceApiClient
14
+
15
+ _client = None
16
+
17
+
18
+ def get_client():
19
+ """Get or create a singleton API client instance."""
20
+ global _client
21
+ if _client is None:
22
+ base_url = os.getenv("DATA_SCIENCE_MCP_URL", "http://localhost:8080")
23
+ token = os.getenv("DATA_SCIENCE_MCP_TOKEN", "")
24
+ verify = os.getenv("DATA_SCIENCE_MCP_VERIFY", "True").lower() in ("true", "1", "yes")
25
+
26
+ try:
27
+ # TODO: Uncomment and configure once the API wrapper class is created
28
+ # _client = DataScienceApiClient(
29
+ # base_url=base_url,
30
+ # token=token,
31
+ # verify=verify,
32
+ # )
33
+
34
+ # Placeholder until API wrapper is implemented
35
+ if _client is None:
36
+ session = requests.Session()
37
+ session.headers.update({"Authorization": f"Bearer {token}"})
38
+ session.verify = verify
39
+ _client = type("Client", (), {"session": session, "base_url": base_url})()
40
+ except (AuthError, UnauthorizedError) as e:
41
+ raise RuntimeError(
42
+ f"AUTHENTICATION ERROR: The credentials provided are not valid for '{base_url}'. "
43
+ f"Please check your DATA_SCIENCE_MCP_TOKEN and DATA_SCIENCE_MCP_URL environment variables. "
44
+ f"Error details: {str(e)}"
45
+ ) from e
46
+
47
+ return _client