langchain-mcp-tools 0.0.1__tar.gz → 0.0.3__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.
- langchain_mcp_tools-0.0.3/PKG-INFO +104 -0
- langchain_mcp_tools-0.0.3/README.md +83 -0
- langchain_mcp_tools-0.0.3/langchain_mcp_tools.egg-info/PKG-INFO +104 -0
- langchain_mcp_tools-0.0.3/langchain_mcp_tools.egg-info/SOURCES.txt +10 -0
- {langchain_mcp_tools-0.0.1/src → langchain_mcp_tools-0.0.3}/langchain_mcp_tools.egg-info/requires.txt +1 -0
- langchain_mcp_tools-0.0.3/langchain_mcp_tools.egg-info/top_level.txt +2 -0
- {langchain_mcp_tools-0.0.1 → langchain_mcp_tools-0.0.3}/pyproject.toml +14 -2
- langchain_mcp_tools-0.0.1/PKG-INFO +0 -93
- langchain_mcp_tools-0.0.1/README.md +0 -75
- langchain_mcp_tools-0.0.1/src/cli_chat.py +0 -274
- langchain_mcp_tools-0.0.1/src/config_loader.py +0 -39
- langchain_mcp_tools-0.0.1/src/langchain_mcp_tools.egg-info/PKG-INFO +0 -93
- langchain_mcp_tools-0.0.1/src/langchain_mcp_tools.egg-info/SOURCES.txt +0 -12
- langchain_mcp_tools-0.0.1/src/langchain_mcp_tools.egg-info/top_level.txt +0 -3
- {langchain_mcp_tools-0.0.1 → langchain_mcp_tools-0.0.3}/LICENSE +0 -0
- {langchain_mcp_tools-0.0.1/src → langchain_mcp_tools-0.0.3}/langchain_mcp_tools.egg-info/dependency_links.txt +0 -0
- {langchain_mcp_tools-0.0.1 → langchain_mcp_tools-0.0.3}/setup.cfg +0 -0
- {langchain_mcp_tools-0.0.1 → langchain_mcp_tools-0.0.3}/src/langchain_mcp_tools.py +0 -0
- {langchain_mcp_tools-0.0.1 → langchain_mcp_tools-0.0.3}/tests/test_langchain_mcp_tools.py +0 -0
|
@@ -0,0 +1,104 @@
|
|
|
1
|
+
Metadata-Version: 2.2
|
|
2
|
+
Name: langchain-mcp-tools
|
|
3
|
+
Version: 0.0.3
|
|
4
|
+
Summary: MCP To LangChain Tools Conversion Utility
|
|
5
|
+
Project-URL: Bug Tracker, https://github.com/hideya/langchain-mcp-tools-py/issues
|
|
6
|
+
Project-URL: Source Code, https://github.com/hideya/langchain-mcp-tools-py
|
|
7
|
+
Requires-Python: >=3.11
|
|
8
|
+
Description-Content-Type: text/markdown
|
|
9
|
+
License-File: LICENSE
|
|
10
|
+
Requires-Dist: jsonschema-pydantic>=0.6
|
|
11
|
+
Requires-Dist: langchain-core>=0.3.29
|
|
12
|
+
Requires-Dist: langchain>=0.3.14
|
|
13
|
+
Requires-Dist: langchain-anthropic>=0.3.1
|
|
14
|
+
Requires-Dist: langchain-groq>=0.2.3
|
|
15
|
+
Requires-Dist: langchain-openai>=0.3.0
|
|
16
|
+
Requires-Dist: langgraph>=0.2.62
|
|
17
|
+
Requires-Dist: mcp>=1.2.0
|
|
18
|
+
Requires-Dist: pyjson5>=1.6.8
|
|
19
|
+
Requires-Dist: pympler>=1.1
|
|
20
|
+
Requires-Dist: python-dotenv>=1.0.1
|
|
21
|
+
|
|
22
|
+
# MCP To LangChain Tools Conversion Utility / Python [](https://github.com/hideya/mcp-langchain-client-ts/blob/main/LICENSE) [](https://pypi.org/project/langchain-mcp-tools/)
|
|
23
|
+
|
|
24
|
+
This package is intended to simplify the use of
|
|
25
|
+
[Model Context Protocol (MCP)](https://modelcontextprotocol.io/)
|
|
26
|
+
server tools with LangChain / Python.
|
|
27
|
+
|
|
28
|
+
It contains a utility function `convertMcpToLangchainTools()`.
|
|
29
|
+
This function handles parallel initialization of specified multiple MCP servers
|
|
30
|
+
and converts their available tools into an array of
|
|
31
|
+
[LangChain-compatible tools](https://js.langchain.com/docs/how_to/tool_calling/).
|
|
32
|
+
|
|
33
|
+
A typescript version of this MCP client is available
|
|
34
|
+
[here](https://www.npmjs.com/package/@h1deya/langchain-mcp-tools)
|
|
35
|
+
|
|
36
|
+
## Requirements
|
|
37
|
+
|
|
38
|
+
- Python 3.11+
|
|
39
|
+
|
|
40
|
+
## Installation
|
|
41
|
+
|
|
42
|
+
```bash
|
|
43
|
+
pip install langchain-mcp-tools
|
|
44
|
+
```
|
|
45
|
+
|
|
46
|
+
## Quick Start
|
|
47
|
+
|
|
48
|
+
`convert_mcp_to_langchain_tools()` utility function accepts MCP server configurations
|
|
49
|
+
that follow the same structure as
|
|
50
|
+
[Claude for Desktop](https://modelcontextprotocol.io/quickstart/user),
|
|
51
|
+
but only the contents of the `mcpServers` property,
|
|
52
|
+
and is expressed as a `dict`, e.g.:
|
|
53
|
+
|
|
54
|
+
```python
|
|
55
|
+
mcp_configs = {
|
|
56
|
+
'filesystem': {
|
|
57
|
+
'command': 'npx',
|
|
58
|
+
'args': ['-y', '@modelcontextprotocol/server-filesystem', '.']
|
|
59
|
+
},
|
|
60
|
+
'fetch': {
|
|
61
|
+
'command': 'uvx',
|
|
62
|
+
'args': ['mcp-server-fetch']
|
|
63
|
+
}
|
|
64
|
+
}
|
|
65
|
+
|
|
66
|
+
tools, mcp_cleanup = await convert_mcp_to_langchain_tools(
|
|
67
|
+
mcp_configs
|
|
68
|
+
)
|
|
69
|
+
```
|
|
70
|
+
|
|
71
|
+
The utility function initializes all specified MCP servers in parallel,
|
|
72
|
+
and returns LangChain Tools (`List[BaseTool]`)
|
|
73
|
+
by gathering all available MCP server tools,
|
|
74
|
+
and by wrapping them into [LangChain Tools](https://js.langchain.com/docs/how_to/tool_calling/).
|
|
75
|
+
It also returns a cleanup callback function (`McpServerCleanupFn`)
|
|
76
|
+
to be invoked to close all MCP server sessions when finished.
|
|
77
|
+
|
|
78
|
+
The returned tools can be used with LangChain, e.g.:
|
|
79
|
+
|
|
80
|
+
```python
|
|
81
|
+
# from langchain.chat_models import init_chat_model
|
|
82
|
+
llm = init_chat_model(
|
|
83
|
+
model='claude-3-5-haiku-latest',
|
|
84
|
+
model_provider='anthropic'
|
|
85
|
+
)
|
|
86
|
+
|
|
87
|
+
# from langgraph.prebuilt import create_react_agent
|
|
88
|
+
agent = create_react_agent(
|
|
89
|
+
llm,
|
|
90
|
+
tools
|
|
91
|
+
)
|
|
92
|
+
```
|
|
93
|
+
<!-- A simple and experimentable usage example can be found
|
|
94
|
+
[here](https://github.com/hideya/langchain-mcp-tools-ts-usage/blob/main/src/index.ts) -->
|
|
95
|
+
|
|
96
|
+
<!-- A more realistic usage example can be found
|
|
97
|
+
[here](https://github.com/hideya/langchain-mcp-client-ts) -->
|
|
98
|
+
|
|
99
|
+
An usage example can be found
|
|
100
|
+
[here](https://github.com/hideya/mcp-client-langchain-py)
|
|
101
|
+
|
|
102
|
+
## Limitations
|
|
103
|
+
|
|
104
|
+
Currently, only text results of tool calls are supported.
|
|
@@ -0,0 +1,83 @@
|
|
|
1
|
+
# MCP To LangChain Tools Conversion Utility / Python [](https://github.com/hideya/mcp-langchain-client-ts/blob/main/LICENSE) [](https://pypi.org/project/langchain-mcp-tools/)
|
|
2
|
+
|
|
3
|
+
This package is intended to simplify the use of
|
|
4
|
+
[Model Context Protocol (MCP)](https://modelcontextprotocol.io/)
|
|
5
|
+
server tools with LangChain / Python.
|
|
6
|
+
|
|
7
|
+
It contains a utility function `convertMcpToLangchainTools()`.
|
|
8
|
+
This function handles parallel initialization of specified multiple MCP servers
|
|
9
|
+
and converts their available tools into an array of
|
|
10
|
+
[LangChain-compatible tools](https://js.langchain.com/docs/how_to/tool_calling/).
|
|
11
|
+
|
|
12
|
+
A typescript version of this MCP client is available
|
|
13
|
+
[here](https://www.npmjs.com/package/@h1deya/langchain-mcp-tools)
|
|
14
|
+
|
|
15
|
+
## Requirements
|
|
16
|
+
|
|
17
|
+
- Python 3.11+
|
|
18
|
+
|
|
19
|
+
## Installation
|
|
20
|
+
|
|
21
|
+
```bash
|
|
22
|
+
pip install langchain-mcp-tools
|
|
23
|
+
```
|
|
24
|
+
|
|
25
|
+
## Quick Start
|
|
26
|
+
|
|
27
|
+
`convert_mcp_to_langchain_tools()` utility function accepts MCP server configurations
|
|
28
|
+
that follow the same structure as
|
|
29
|
+
[Claude for Desktop](https://modelcontextprotocol.io/quickstart/user),
|
|
30
|
+
but only the contents of the `mcpServers` property,
|
|
31
|
+
and is expressed as a `dict`, e.g.:
|
|
32
|
+
|
|
33
|
+
```python
|
|
34
|
+
mcp_configs = {
|
|
35
|
+
'filesystem': {
|
|
36
|
+
'command': 'npx',
|
|
37
|
+
'args': ['-y', '@modelcontextprotocol/server-filesystem', '.']
|
|
38
|
+
},
|
|
39
|
+
'fetch': {
|
|
40
|
+
'command': 'uvx',
|
|
41
|
+
'args': ['mcp-server-fetch']
|
|
42
|
+
}
|
|
43
|
+
}
|
|
44
|
+
|
|
45
|
+
tools, mcp_cleanup = await convert_mcp_to_langchain_tools(
|
|
46
|
+
mcp_configs
|
|
47
|
+
)
|
|
48
|
+
```
|
|
49
|
+
|
|
50
|
+
The utility function initializes all specified MCP servers in parallel,
|
|
51
|
+
and returns LangChain Tools (`List[BaseTool]`)
|
|
52
|
+
by gathering all available MCP server tools,
|
|
53
|
+
and by wrapping them into [LangChain Tools](https://js.langchain.com/docs/how_to/tool_calling/).
|
|
54
|
+
It also returns a cleanup callback function (`McpServerCleanupFn`)
|
|
55
|
+
to be invoked to close all MCP server sessions when finished.
|
|
56
|
+
|
|
57
|
+
The returned tools can be used with LangChain, e.g.:
|
|
58
|
+
|
|
59
|
+
```python
|
|
60
|
+
# from langchain.chat_models import init_chat_model
|
|
61
|
+
llm = init_chat_model(
|
|
62
|
+
model='claude-3-5-haiku-latest',
|
|
63
|
+
model_provider='anthropic'
|
|
64
|
+
)
|
|
65
|
+
|
|
66
|
+
# from langgraph.prebuilt import create_react_agent
|
|
67
|
+
agent = create_react_agent(
|
|
68
|
+
llm,
|
|
69
|
+
tools
|
|
70
|
+
)
|
|
71
|
+
```
|
|
72
|
+
<!-- A simple and experimentable usage example can be found
|
|
73
|
+
[here](https://github.com/hideya/langchain-mcp-tools-ts-usage/blob/main/src/index.ts) -->
|
|
74
|
+
|
|
75
|
+
<!-- A more realistic usage example can be found
|
|
76
|
+
[here](https://github.com/hideya/langchain-mcp-client-ts) -->
|
|
77
|
+
|
|
78
|
+
An usage example can be found
|
|
79
|
+
[here](https://github.com/hideya/mcp-client-langchain-py)
|
|
80
|
+
|
|
81
|
+
## Limitations
|
|
82
|
+
|
|
83
|
+
Currently, only text results of tool calls are supported.
|
|
@@ -0,0 +1,104 @@
|
|
|
1
|
+
Metadata-Version: 2.2
|
|
2
|
+
Name: langchain-mcp-tools
|
|
3
|
+
Version: 0.0.3
|
|
4
|
+
Summary: MCP To LangChain Tools Conversion Utility
|
|
5
|
+
Project-URL: Bug Tracker, https://github.com/hideya/langchain-mcp-tools-py/issues
|
|
6
|
+
Project-URL: Source Code, https://github.com/hideya/langchain-mcp-tools-py
|
|
7
|
+
Requires-Python: >=3.11
|
|
8
|
+
Description-Content-Type: text/markdown
|
|
9
|
+
License-File: LICENSE
|
|
10
|
+
Requires-Dist: jsonschema-pydantic>=0.6
|
|
11
|
+
Requires-Dist: langchain-core>=0.3.29
|
|
12
|
+
Requires-Dist: langchain>=0.3.14
|
|
13
|
+
Requires-Dist: langchain-anthropic>=0.3.1
|
|
14
|
+
Requires-Dist: langchain-groq>=0.2.3
|
|
15
|
+
Requires-Dist: langchain-openai>=0.3.0
|
|
16
|
+
Requires-Dist: langgraph>=0.2.62
|
|
17
|
+
Requires-Dist: mcp>=1.2.0
|
|
18
|
+
Requires-Dist: pyjson5>=1.6.8
|
|
19
|
+
Requires-Dist: pympler>=1.1
|
|
20
|
+
Requires-Dist: python-dotenv>=1.0.1
|
|
21
|
+
|
|
22
|
+
# MCP To LangChain Tools Conversion Utility / Python [](https://github.com/hideya/mcp-langchain-client-ts/blob/main/LICENSE) [](https://pypi.org/project/langchain-mcp-tools/)
|
|
23
|
+
|
|
24
|
+
This package is intended to simplify the use of
|
|
25
|
+
[Model Context Protocol (MCP)](https://modelcontextprotocol.io/)
|
|
26
|
+
server tools with LangChain / Python.
|
|
27
|
+
|
|
28
|
+
It contains a utility function `convertMcpToLangchainTools()`.
|
|
29
|
+
This function handles parallel initialization of specified multiple MCP servers
|
|
30
|
+
and converts their available tools into an array of
|
|
31
|
+
[LangChain-compatible tools](https://js.langchain.com/docs/how_to/tool_calling/).
|
|
32
|
+
|
|
33
|
+
A typescript version of this MCP client is available
|
|
34
|
+
[here](https://www.npmjs.com/package/@h1deya/langchain-mcp-tools)
|
|
35
|
+
|
|
36
|
+
## Requirements
|
|
37
|
+
|
|
38
|
+
- Python 3.11+
|
|
39
|
+
|
|
40
|
+
## Installation
|
|
41
|
+
|
|
42
|
+
```bash
|
|
43
|
+
pip install langchain-mcp-tools
|
|
44
|
+
```
|
|
45
|
+
|
|
46
|
+
## Quick Start
|
|
47
|
+
|
|
48
|
+
`convert_mcp_to_langchain_tools()` utility function accepts MCP server configurations
|
|
49
|
+
that follow the same structure as
|
|
50
|
+
[Claude for Desktop](https://modelcontextprotocol.io/quickstart/user),
|
|
51
|
+
but only the contents of the `mcpServers` property,
|
|
52
|
+
and is expressed as a `dict`, e.g.:
|
|
53
|
+
|
|
54
|
+
```python
|
|
55
|
+
mcp_configs = {
|
|
56
|
+
'filesystem': {
|
|
57
|
+
'command': 'npx',
|
|
58
|
+
'args': ['-y', '@modelcontextprotocol/server-filesystem', '.']
|
|
59
|
+
},
|
|
60
|
+
'fetch': {
|
|
61
|
+
'command': 'uvx',
|
|
62
|
+
'args': ['mcp-server-fetch']
|
|
63
|
+
}
|
|
64
|
+
}
|
|
65
|
+
|
|
66
|
+
tools, mcp_cleanup = await convert_mcp_to_langchain_tools(
|
|
67
|
+
mcp_configs
|
|
68
|
+
)
|
|
69
|
+
```
|
|
70
|
+
|
|
71
|
+
The utility function initializes all specified MCP servers in parallel,
|
|
72
|
+
and returns LangChain Tools (`List[BaseTool]`)
|
|
73
|
+
by gathering all available MCP server tools,
|
|
74
|
+
and by wrapping them into [LangChain Tools](https://js.langchain.com/docs/how_to/tool_calling/).
|
|
75
|
+
It also returns a cleanup callback function (`McpServerCleanupFn`)
|
|
76
|
+
to be invoked to close all MCP server sessions when finished.
|
|
77
|
+
|
|
78
|
+
The returned tools can be used with LangChain, e.g.:
|
|
79
|
+
|
|
80
|
+
```python
|
|
81
|
+
# from langchain.chat_models import init_chat_model
|
|
82
|
+
llm = init_chat_model(
|
|
83
|
+
model='claude-3-5-haiku-latest',
|
|
84
|
+
model_provider='anthropic'
|
|
85
|
+
)
|
|
86
|
+
|
|
87
|
+
# from langgraph.prebuilt import create_react_agent
|
|
88
|
+
agent = create_react_agent(
|
|
89
|
+
llm,
|
|
90
|
+
tools
|
|
91
|
+
)
|
|
92
|
+
```
|
|
93
|
+
<!-- A simple and experimentable usage example can be found
|
|
94
|
+
[here](https://github.com/hideya/langchain-mcp-tools-ts-usage/blob/main/src/index.ts) -->
|
|
95
|
+
|
|
96
|
+
<!-- A more realistic usage example can be found
|
|
97
|
+
[here](https://github.com/hideya/langchain-mcp-client-ts) -->
|
|
98
|
+
|
|
99
|
+
An usage example can be found
|
|
100
|
+
[here](https://github.com/hideya/mcp-client-langchain-py)
|
|
101
|
+
|
|
102
|
+
## Limitations
|
|
103
|
+
|
|
104
|
+
Currently, only text results of tool calls are supported.
|
|
@@ -0,0 +1,10 @@
|
|
|
1
|
+
LICENSE
|
|
2
|
+
README.md
|
|
3
|
+
pyproject.toml
|
|
4
|
+
langchain_mcp_tools.egg-info/PKG-INFO
|
|
5
|
+
langchain_mcp_tools.egg-info/SOURCES.txt
|
|
6
|
+
langchain_mcp_tools.egg-info/dependency_links.txt
|
|
7
|
+
langchain_mcp_tools.egg-info/requires.txt
|
|
8
|
+
langchain_mcp_tools.egg-info/top_level.txt
|
|
9
|
+
src/langchain_mcp_tools.py
|
|
10
|
+
tests/test_langchain_mcp_tools.py
|
|
@@ -1,11 +1,12 @@
|
|
|
1
1
|
[project]
|
|
2
2
|
name = "langchain-mcp-tools"
|
|
3
|
-
version = "0.0.
|
|
4
|
-
description = "MCP
|
|
3
|
+
version = "0.0.3"
|
|
4
|
+
description = "MCP To LangChain Tools Conversion Utility"
|
|
5
5
|
readme = "README.md"
|
|
6
6
|
requires-python = ">=3.11"
|
|
7
7
|
dependencies = [
|
|
8
8
|
"jsonschema-pydantic>=0.6",
|
|
9
|
+
"langchain-core>=0.3.29",
|
|
9
10
|
"langchain>=0.3.14",
|
|
10
11
|
"langchain-anthropic>=0.3.1",
|
|
11
12
|
"langchain-groq>=0.2.3",
|
|
@@ -23,3 +24,14 @@ dev = [
|
|
|
23
24
|
"pytest-asyncio>=0.25.2",
|
|
24
25
|
"twine>=6.0.1",
|
|
25
26
|
]
|
|
27
|
+
|
|
28
|
+
[tool.setuptools.packages.find]
|
|
29
|
+
exclude = [
|
|
30
|
+
"tests*",
|
|
31
|
+
"docs*",
|
|
32
|
+
"examples*",
|
|
33
|
+
]
|
|
34
|
+
|
|
35
|
+
[project.urls]
|
|
36
|
+
"Bug Tracker" = "https://github.com/hideya/langchain-mcp-tools-py/issues"
|
|
37
|
+
"Source Code" = "https://github.com/hideya/langchain-mcp-tools-py"
|
|
@@ -1,93 +0,0 @@
|
|
|
1
|
-
Metadata-Version: 2.2
|
|
2
|
-
Name: langchain-mcp-tools
|
|
3
|
-
Version: 0.0.1
|
|
4
|
-
Summary: MCP Client Using LangChain / Python
|
|
5
|
-
Requires-Python: >=3.11
|
|
6
|
-
Description-Content-Type: text/markdown
|
|
7
|
-
License-File: LICENSE
|
|
8
|
-
Requires-Dist: jsonschema-pydantic>=0.6
|
|
9
|
-
Requires-Dist: langchain>=0.3.14
|
|
10
|
-
Requires-Dist: langchain-anthropic>=0.3.1
|
|
11
|
-
Requires-Dist: langchain-groq>=0.2.3
|
|
12
|
-
Requires-Dist: langchain-openai>=0.3.0
|
|
13
|
-
Requires-Dist: langgraph>=0.2.62
|
|
14
|
-
Requires-Dist: mcp>=1.2.0
|
|
15
|
-
Requires-Dist: pyjson5>=1.6.8
|
|
16
|
-
Requires-Dist: pympler>=1.1
|
|
17
|
-
Requires-Dist: python-dotenv>=1.0.1
|
|
18
|
-
|
|
19
|
-
# MCP Client Using LangChain / Python [](https://github.com/hideya/mcp-langchain-client-ts/blob/main/LICENSE)
|
|
20
|
-
|
|
21
|
-
This simple [Model Context Protocol (MCP)](https://modelcontextprotocol.io/)
|
|
22
|
-
client demonstrates MCP server invocations by LangChain ReAct Agent.
|
|
23
|
-
|
|
24
|
-
It leverages a utility function `convert_mcp_to_langchain_tools()` from
|
|
25
|
-
`langchain_mcp_tools`.
|
|
26
|
-
This function handles parallel initialization of specified multiple MCP servers
|
|
27
|
-
and converts their available tools into a list of
|
|
28
|
-
[LangChain-compatible tools](https://js.langchain.com/docs/how_to/tool_calling/).
|
|
29
|
-
|
|
30
|
-
LLMs from Anthropic, OpenAI and Groq are currently supported.
|
|
31
|
-
|
|
32
|
-
A typescript version of this MCP client is available
|
|
33
|
-
[here](https://github.com/hideya/mcp-client-langchain-ts)
|
|
34
|
-
|
|
35
|
-
## Requirements
|
|
36
|
-
|
|
37
|
-
- Python 3.11+
|
|
38
|
-
- [`uv`](https://docs.astral.sh/uv/) installation
|
|
39
|
-
- API keys from [Anthropic](https://console.anthropic.com/settings/keys),
|
|
40
|
-
[OpenAI](https://platform.openai.com/api-keys), and/or
|
|
41
|
-
[Groq](https://console.groq.com/keys)
|
|
42
|
-
as needed
|
|
43
|
-
|
|
44
|
-
## Setup
|
|
45
|
-
1. Install dependencies:
|
|
46
|
-
```bash
|
|
47
|
-
make install
|
|
48
|
-
```
|
|
49
|
-
|
|
50
|
-
2. Setup API keys:
|
|
51
|
-
```bash
|
|
52
|
-
cp .env.template .env
|
|
53
|
-
```
|
|
54
|
-
- Update `.env` as needed.
|
|
55
|
-
- `.gitignore` is configured to ignore `.env`
|
|
56
|
-
to prevent accidental commits of the credentials.
|
|
57
|
-
|
|
58
|
-
3. Configure LLM and MCP Servers settings `llm_mcp_config.json5` as needed.
|
|
59
|
-
|
|
60
|
-
- [The configuration file format](https://github.com/hideya/mcp-client-langchain-ts/blob/main/llm_mcp_config.json5)
|
|
61
|
-
for MCP servers follows the same structure as
|
|
62
|
-
[Claude for Desktop](https://modelcontextprotocol.io/quickstart/user),
|
|
63
|
-
with one difference: the key name `mcpServers` has been changed
|
|
64
|
-
to `mcp_servers` to follow the snake_case convention
|
|
65
|
-
commonly used in JSON configuration files.
|
|
66
|
-
- The file format is [JSON5](https://json5.org/),
|
|
67
|
-
where comments and trailing commas are allowed.
|
|
68
|
-
- The format is further extended to replace `${...}` notations
|
|
69
|
-
with the values of corresponding environment variables.
|
|
70
|
-
- Keep all the credentials and private info in the `.env` file
|
|
71
|
-
and refer to them with `${...}` notation as needed.
|
|
72
|
-
|
|
73
|
-
|
|
74
|
-
## Usage
|
|
75
|
-
|
|
76
|
-
Run the app:
|
|
77
|
-
```bash
|
|
78
|
-
make start
|
|
79
|
-
```
|
|
80
|
-
|
|
81
|
-
Run in verbose mode:
|
|
82
|
-
```bash
|
|
83
|
-
make start-v
|
|
84
|
-
```
|
|
85
|
-
|
|
86
|
-
See commandline options:
|
|
87
|
-
```bash
|
|
88
|
-
make start-h
|
|
89
|
-
```
|
|
90
|
-
|
|
91
|
-
At the prompt, you can simply press Enter to use example queries that perform MCP server tool invocations.
|
|
92
|
-
|
|
93
|
-
Example queries can be configured in `llm_mcp_config.json5`
|
|
@@ -1,75 +0,0 @@
|
|
|
1
|
-
# MCP Client Using LangChain / Python [](https://github.com/hideya/mcp-langchain-client-ts/blob/main/LICENSE)
|
|
2
|
-
|
|
3
|
-
This simple [Model Context Protocol (MCP)](https://modelcontextprotocol.io/)
|
|
4
|
-
client demonstrates MCP server invocations by LangChain ReAct Agent.
|
|
5
|
-
|
|
6
|
-
It leverages a utility function `convert_mcp_to_langchain_tools()` from
|
|
7
|
-
`langchain_mcp_tools`.
|
|
8
|
-
This function handles parallel initialization of specified multiple MCP servers
|
|
9
|
-
and converts their available tools into a list of
|
|
10
|
-
[LangChain-compatible tools](https://js.langchain.com/docs/how_to/tool_calling/).
|
|
11
|
-
|
|
12
|
-
LLMs from Anthropic, OpenAI and Groq are currently supported.
|
|
13
|
-
|
|
14
|
-
A typescript version of this MCP client is available
|
|
15
|
-
[here](https://github.com/hideya/mcp-client-langchain-ts)
|
|
16
|
-
|
|
17
|
-
## Requirements
|
|
18
|
-
|
|
19
|
-
- Python 3.11+
|
|
20
|
-
- [`uv`](https://docs.astral.sh/uv/) installation
|
|
21
|
-
- API keys from [Anthropic](https://console.anthropic.com/settings/keys),
|
|
22
|
-
[OpenAI](https://platform.openai.com/api-keys), and/or
|
|
23
|
-
[Groq](https://console.groq.com/keys)
|
|
24
|
-
as needed
|
|
25
|
-
|
|
26
|
-
## Setup
|
|
27
|
-
1. Install dependencies:
|
|
28
|
-
```bash
|
|
29
|
-
make install
|
|
30
|
-
```
|
|
31
|
-
|
|
32
|
-
2. Setup API keys:
|
|
33
|
-
```bash
|
|
34
|
-
cp .env.template .env
|
|
35
|
-
```
|
|
36
|
-
- Update `.env` as needed.
|
|
37
|
-
- `.gitignore` is configured to ignore `.env`
|
|
38
|
-
to prevent accidental commits of the credentials.
|
|
39
|
-
|
|
40
|
-
3. Configure LLM and MCP Servers settings `llm_mcp_config.json5` as needed.
|
|
41
|
-
|
|
42
|
-
- [The configuration file format](https://github.com/hideya/mcp-client-langchain-ts/blob/main/llm_mcp_config.json5)
|
|
43
|
-
for MCP servers follows the same structure as
|
|
44
|
-
[Claude for Desktop](https://modelcontextprotocol.io/quickstart/user),
|
|
45
|
-
with one difference: the key name `mcpServers` has been changed
|
|
46
|
-
to `mcp_servers` to follow the snake_case convention
|
|
47
|
-
commonly used in JSON configuration files.
|
|
48
|
-
- The file format is [JSON5](https://json5.org/),
|
|
49
|
-
where comments and trailing commas are allowed.
|
|
50
|
-
- The format is further extended to replace `${...}` notations
|
|
51
|
-
with the values of corresponding environment variables.
|
|
52
|
-
- Keep all the credentials and private info in the `.env` file
|
|
53
|
-
and refer to them with `${...}` notation as needed.
|
|
54
|
-
|
|
55
|
-
|
|
56
|
-
## Usage
|
|
57
|
-
|
|
58
|
-
Run the app:
|
|
59
|
-
```bash
|
|
60
|
-
make start
|
|
61
|
-
```
|
|
62
|
-
|
|
63
|
-
Run in verbose mode:
|
|
64
|
-
```bash
|
|
65
|
-
make start-v
|
|
66
|
-
```
|
|
67
|
-
|
|
68
|
-
See commandline options:
|
|
69
|
-
```bash
|
|
70
|
-
make start-h
|
|
71
|
-
```
|
|
72
|
-
|
|
73
|
-
At the prompt, you can simply press Enter to use example queries that perform MCP server tool invocations.
|
|
74
|
-
|
|
75
|
-
Example queries can be configured in `llm_mcp_config.json5`
|
|
@@ -1,274 +0,0 @@
|
|
|
1
|
-
# Standard library imports
|
|
2
|
-
import argparse
|
|
3
|
-
import asyncio
|
|
4
|
-
from enum import Enum
|
|
5
|
-
import json
|
|
6
|
-
import logging
|
|
7
|
-
import sys
|
|
8
|
-
from pathlib import Path
|
|
9
|
-
from typing import (
|
|
10
|
-
List,
|
|
11
|
-
Optional,
|
|
12
|
-
Dict,
|
|
13
|
-
Any,
|
|
14
|
-
cast,
|
|
15
|
-
)
|
|
16
|
-
|
|
17
|
-
# Third-party imports
|
|
18
|
-
try:
|
|
19
|
-
from dotenv import load_dotenv
|
|
20
|
-
from langchain.chat_models import init_chat_model
|
|
21
|
-
from langchain.schema import (
|
|
22
|
-
AIMessage,
|
|
23
|
-
BaseMessage,
|
|
24
|
-
HumanMessage,
|
|
25
|
-
SystemMessage,
|
|
26
|
-
)
|
|
27
|
-
from langchain_core.runnables.base import Runnable
|
|
28
|
-
from langchain_core.messages.tool import ToolMessage
|
|
29
|
-
from langgraph.prebuilt import create_react_agent
|
|
30
|
-
except ImportError as e:
|
|
31
|
-
print(f'\nError: Required package not found: {e}')
|
|
32
|
-
print('Please ensure all required packages are installed\n')
|
|
33
|
-
sys.exit(1)
|
|
34
|
-
|
|
35
|
-
# Local application imports
|
|
36
|
-
from config_loader import load_config
|
|
37
|
-
from langchain_mcp_tools import (
|
|
38
|
-
convert_mcp_to_langchain_tools,
|
|
39
|
-
McpServerCleanupFn,
|
|
40
|
-
)
|
|
41
|
-
|
|
42
|
-
# Type definitions
|
|
43
|
-
ConfigType = Dict[str, Any]
|
|
44
|
-
|
|
45
|
-
|
|
46
|
-
# ANSI color escape codes
|
|
47
|
-
class Colors(str, Enum):
|
|
48
|
-
YELLOW = '\033[33m' # color to yellow
|
|
49
|
-
CYAN = '\033[36m' # color to cyan
|
|
50
|
-
RESET = '\033[0m' # reset color
|
|
51
|
-
|
|
52
|
-
def __str__(self):
|
|
53
|
-
return self.value
|
|
54
|
-
|
|
55
|
-
|
|
56
|
-
def parse_arguments() -> argparse.Namespace:
|
|
57
|
-
"""Parse and return command line args for config path and verbosity."""
|
|
58
|
-
parser = argparse.ArgumentParser(
|
|
59
|
-
description='CLI Chat Application',
|
|
60
|
-
formatter_class=argparse.ArgumentDefaultsHelpFormatter
|
|
61
|
-
)
|
|
62
|
-
parser.add_argument(
|
|
63
|
-
'-c', '--config',
|
|
64
|
-
default='llm_mcp_config.json5',
|
|
65
|
-
help='path to config file',
|
|
66
|
-
type=Path,
|
|
67
|
-
metavar='PATH'
|
|
68
|
-
)
|
|
69
|
-
parser.add_argument(
|
|
70
|
-
'-v', '--verbose',
|
|
71
|
-
action='store_true',
|
|
72
|
-
help='run with verbose logging'
|
|
73
|
-
)
|
|
74
|
-
return parser.parse_args()
|
|
75
|
-
|
|
76
|
-
|
|
77
|
-
def init_logger(verbose: bool) -> logging.Logger:
|
|
78
|
-
"""Initialize and return a logger with appropriate verbosity level."""
|
|
79
|
-
logging.basicConfig(
|
|
80
|
-
level=logging.DEBUG if verbose else logging.INFO,
|
|
81
|
-
format='\x1b[90m[%(levelname)s]\x1b[0m %(message)s'
|
|
82
|
-
)
|
|
83
|
-
return logging.getLogger()
|
|
84
|
-
|
|
85
|
-
|
|
86
|
-
def print_colored(text: str, color: Colors, end: str = "\n") -> None:
|
|
87
|
-
"""Print text in specified color and reset afterwards."""
|
|
88
|
-
print(f"{color}{text}{Colors.RESET}", end=end)
|
|
89
|
-
|
|
90
|
-
|
|
91
|
-
def set_color(color: Colors) -> None:
|
|
92
|
-
"""Set terminal color."""
|
|
93
|
-
print(color, end='')
|
|
94
|
-
|
|
95
|
-
|
|
96
|
-
def clear_line() -> None:
|
|
97
|
-
"""Move up one line and clear it."""
|
|
98
|
-
print('\x1b[1A\x1b[2K', end='')
|
|
99
|
-
|
|
100
|
-
|
|
101
|
-
async def get_user_query(remaining_queries: List[str]) -> Optional[str]:
|
|
102
|
-
"""Get user input or next example query, handling empty inputs
|
|
103
|
-
and quit commands."""
|
|
104
|
-
set_color(Colors.YELLOW)
|
|
105
|
-
query = input('Query: ').strip()
|
|
106
|
-
|
|
107
|
-
if len(query) == 0:
|
|
108
|
-
if len(remaining_queries) > 0:
|
|
109
|
-
query = remaining_queries.pop(0)
|
|
110
|
-
clear_line()
|
|
111
|
-
print_colored(f'Example Query: {query}', Colors.YELLOW)
|
|
112
|
-
else:
|
|
113
|
-
set_color(Colors.RESET)
|
|
114
|
-
print('\nPlease type a query, or "quit" or "q" to exit\n')
|
|
115
|
-
return await get_user_query(remaining_queries)
|
|
116
|
-
|
|
117
|
-
print(Colors.RESET) # Reset after input
|
|
118
|
-
|
|
119
|
-
if query.lower() in ['quit', 'q']:
|
|
120
|
-
print_colored('Goodbye!\n', Colors.CYAN)
|
|
121
|
-
return None
|
|
122
|
-
|
|
123
|
-
return query
|
|
124
|
-
|
|
125
|
-
|
|
126
|
-
async def handle_conversation(
|
|
127
|
-
agent: Runnable,
|
|
128
|
-
messages: List[BaseMessage],
|
|
129
|
-
example_queries: List[str],
|
|
130
|
-
verbose: bool
|
|
131
|
-
) -> None:
|
|
132
|
-
"""Manage an interactive conversation loop between the user and AI agent.
|
|
133
|
-
|
|
134
|
-
Args:
|
|
135
|
-
agent (Runnable): The initialized ReAct agent that processes queries
|
|
136
|
-
messages (List[BaseMessage]): List to maintain conversation history
|
|
137
|
-
example_queries (List[str]): list of example queries that can be used
|
|
138
|
-
when user presses Enter
|
|
139
|
-
verbose (bool): Flag to control detailed output of tool responses
|
|
140
|
-
|
|
141
|
-
Exception handling:
|
|
142
|
-
- TypeError: Ensures response is in correct string format
|
|
143
|
-
- General exceptions: Allows conversation to continue after errors
|
|
144
|
-
|
|
145
|
-
The conversation continues until user types 'quit' or 'q'.
|
|
146
|
-
"""
|
|
147
|
-
print('\nConversation started. '
|
|
148
|
-
'Type "quit" or "q" to end the conversation.\n')
|
|
149
|
-
if len(example_queries) > 0:
|
|
150
|
-
print('Example Queries (just type Enter to supply them one by one):')
|
|
151
|
-
for ex_q in example_queries:
|
|
152
|
-
print(f"- {ex_q}")
|
|
153
|
-
print()
|
|
154
|
-
|
|
155
|
-
while True:
|
|
156
|
-
try:
|
|
157
|
-
query = await get_user_query(example_queries)
|
|
158
|
-
if not query:
|
|
159
|
-
break
|
|
160
|
-
|
|
161
|
-
messages.append(HumanMessage(content=query))
|
|
162
|
-
|
|
163
|
-
result = await agent.ainvoke({
|
|
164
|
-
'messages': messages
|
|
165
|
-
})
|
|
166
|
-
|
|
167
|
-
result_messages = cast(List[BaseMessage], result['messages'])
|
|
168
|
-
# the last message should be an AIMessage
|
|
169
|
-
response = result_messages[-1].content
|
|
170
|
-
if not isinstance(response, str):
|
|
171
|
-
raise TypeError(
|
|
172
|
-
f"Expected string response, got {type(response)}"
|
|
173
|
-
)
|
|
174
|
-
|
|
175
|
-
# check if msg one before is a ToolMessage
|
|
176
|
-
message_one_before = result_messages[-2]
|
|
177
|
-
if isinstance(message_one_before, ToolMessage):
|
|
178
|
-
if verbose:
|
|
179
|
-
# show tools call response
|
|
180
|
-
print(message_one_before.content)
|
|
181
|
-
# new line after tool call output
|
|
182
|
-
print()
|
|
183
|
-
print_colored(f"{response}\n", Colors.CYAN)
|
|
184
|
-
messages.append(AIMessage(content=response))
|
|
185
|
-
|
|
186
|
-
except Exception as e:
|
|
187
|
-
print(f'Error getting response: {str(e)}')
|
|
188
|
-
print('You can continue chatting or type "quit" to exit.')
|
|
189
|
-
|
|
190
|
-
|
|
191
|
-
async def init_react_agent(
|
|
192
|
-
config: ConfigType,
|
|
193
|
-
logger: logging.Logger
|
|
194
|
-
) -> tuple[Runnable, List[BaseMessage], McpServerCleanupFn]:
|
|
195
|
-
"""Initialize and configure a ReAct agent for conversation handling.
|
|
196
|
-
|
|
197
|
-
Args:
|
|
198
|
-
config (ConfigType): Configuration dictionary containing LLM and
|
|
199
|
-
MCP server settings
|
|
200
|
-
logger (logging.Logger): Logger instance for initialization
|
|
201
|
-
status updates
|
|
202
|
-
|
|
203
|
-
Returns:
|
|
204
|
-
tuple[Runnable, List[BaseMessage], McpServerCleanupFn]:
|
|
205
|
-
Returns a tuple containing:
|
|
206
|
-
- Configured ReAct agent ready for conversation
|
|
207
|
-
- Initial message list (empty or with system prompt)
|
|
208
|
-
- Cleanup function for MCP server connections
|
|
209
|
-
"""
|
|
210
|
-
llm_config = config['llm']
|
|
211
|
-
logger.info(f'Initializing model... {json.dumps(llm_config, indent=2)}\n')
|
|
212
|
-
|
|
213
|
-
llm = init_chat_model(
|
|
214
|
-
model=llm_config['model'],
|
|
215
|
-
model_provider=llm_config['model_provider'],
|
|
216
|
-
temperature=llm_config['temperature'],
|
|
217
|
-
max_tokens=llm_config['max_tokens'],
|
|
218
|
-
)
|
|
219
|
-
|
|
220
|
-
mcp_configs = config['mcp_servers']
|
|
221
|
-
logger.info(f'Initializing {len(mcp_configs)} MCP server(s)...\n')
|
|
222
|
-
tools, mcp_cleanup = await convert_mcp_to_langchain_tools(
|
|
223
|
-
mcp_configs,
|
|
224
|
-
logger
|
|
225
|
-
)
|
|
226
|
-
|
|
227
|
-
agent = create_react_agent(
|
|
228
|
-
llm,
|
|
229
|
-
tools
|
|
230
|
-
)
|
|
231
|
-
|
|
232
|
-
messages: List[BaseMessage] = []
|
|
233
|
-
system_prompt = llm_config.get('system_prompt')
|
|
234
|
-
if system_prompt and isinstance(system_prompt, str):
|
|
235
|
-
messages.append(SystemMessage(content=system_prompt))
|
|
236
|
-
|
|
237
|
-
return agent, messages, mcp_cleanup
|
|
238
|
-
|
|
239
|
-
|
|
240
|
-
async def run() -> None:
|
|
241
|
-
"""Main async function to set up and run the simple chat app."""
|
|
242
|
-
mcp_cleanup: Optional[McpServerCleanupFn] = None
|
|
243
|
-
try:
|
|
244
|
-
load_dotenv()
|
|
245
|
-
args = parse_arguments()
|
|
246
|
-
logger = init_logger(args.verbose)
|
|
247
|
-
config = load_config(args.config)
|
|
248
|
-
example_queries = (
|
|
249
|
-
config.get('example_queries')[:]
|
|
250
|
-
if config.get('example_queries') is not None
|
|
251
|
-
else []
|
|
252
|
-
)
|
|
253
|
-
|
|
254
|
-
agent, messages, mcp_cleanup = await init_react_agent(config, logger)
|
|
255
|
-
|
|
256
|
-
await handle_conversation(
|
|
257
|
-
agent,
|
|
258
|
-
messages,
|
|
259
|
-
example_queries,
|
|
260
|
-
args.verbose
|
|
261
|
-
)
|
|
262
|
-
|
|
263
|
-
finally:
|
|
264
|
-
if mcp_cleanup is not None:
|
|
265
|
-
await mcp_cleanup()
|
|
266
|
-
|
|
267
|
-
|
|
268
|
-
def main() -> None:
|
|
269
|
-
"""Entry point of the script."""
|
|
270
|
-
asyncio.run(run())
|
|
271
|
-
|
|
272
|
-
|
|
273
|
-
if __name__ == '__main__':
|
|
274
|
-
main()
|
|
@@ -1,39 +0,0 @@
|
|
|
1
|
-
import pyjson5 as json5
|
|
2
|
-
from pathlib import Path
|
|
3
|
-
from typing import TypedDict, Optional, Any
|
|
4
|
-
|
|
5
|
-
|
|
6
|
-
class LLMConfig(TypedDict):
|
|
7
|
-
"""Type definition for LLM configuration."""
|
|
8
|
-
model_provider: str
|
|
9
|
-
model: Optional[str]
|
|
10
|
-
temperature: Optional[float]
|
|
11
|
-
system_prompt: Optional[str]
|
|
12
|
-
|
|
13
|
-
|
|
14
|
-
class ConfigError(Exception):
|
|
15
|
-
"""Base exception for configuration related errors."""
|
|
16
|
-
pass
|
|
17
|
-
|
|
18
|
-
|
|
19
|
-
class ConfigFileNotFoundError(ConfigError):
|
|
20
|
-
"""Raised when the configuration file cannot be found."""
|
|
21
|
-
pass
|
|
22
|
-
|
|
23
|
-
|
|
24
|
-
class ConfigValidationError(ConfigError):
|
|
25
|
-
"""Raised when the configuration fails validation."""
|
|
26
|
-
pass
|
|
27
|
-
|
|
28
|
-
|
|
29
|
-
def load_config(config_path: str):
|
|
30
|
-
"""Load and validate configuration from JSON5 file.
|
|
31
|
-
"""
|
|
32
|
-
config_file = Path(config_path)
|
|
33
|
-
if not config_file.exists():
|
|
34
|
-
raise ConfigFileNotFoundError(f"Config file {config_path} not found")
|
|
35
|
-
|
|
36
|
-
with open(config_file, 'r', encoding='utf-8') as f:
|
|
37
|
-
config: dict[str, Any] = json5.load(f)
|
|
38
|
-
|
|
39
|
-
return config
|
|
@@ -1,93 +0,0 @@
|
|
|
1
|
-
Metadata-Version: 2.2
|
|
2
|
-
Name: langchain-mcp-tools
|
|
3
|
-
Version: 0.0.1
|
|
4
|
-
Summary: MCP Client Using LangChain / Python
|
|
5
|
-
Requires-Python: >=3.11
|
|
6
|
-
Description-Content-Type: text/markdown
|
|
7
|
-
License-File: LICENSE
|
|
8
|
-
Requires-Dist: jsonschema-pydantic>=0.6
|
|
9
|
-
Requires-Dist: langchain>=0.3.14
|
|
10
|
-
Requires-Dist: langchain-anthropic>=0.3.1
|
|
11
|
-
Requires-Dist: langchain-groq>=0.2.3
|
|
12
|
-
Requires-Dist: langchain-openai>=0.3.0
|
|
13
|
-
Requires-Dist: langgraph>=0.2.62
|
|
14
|
-
Requires-Dist: mcp>=1.2.0
|
|
15
|
-
Requires-Dist: pyjson5>=1.6.8
|
|
16
|
-
Requires-Dist: pympler>=1.1
|
|
17
|
-
Requires-Dist: python-dotenv>=1.0.1
|
|
18
|
-
|
|
19
|
-
# MCP Client Using LangChain / Python [](https://github.com/hideya/mcp-langchain-client-ts/blob/main/LICENSE)
|
|
20
|
-
|
|
21
|
-
This simple [Model Context Protocol (MCP)](https://modelcontextprotocol.io/)
|
|
22
|
-
client demonstrates MCP server invocations by LangChain ReAct Agent.
|
|
23
|
-
|
|
24
|
-
It leverages a utility function `convert_mcp_to_langchain_tools()` from
|
|
25
|
-
`langchain_mcp_tools`.
|
|
26
|
-
This function handles parallel initialization of specified multiple MCP servers
|
|
27
|
-
and converts their available tools into a list of
|
|
28
|
-
[LangChain-compatible tools](https://js.langchain.com/docs/how_to/tool_calling/).
|
|
29
|
-
|
|
30
|
-
LLMs from Anthropic, OpenAI and Groq are currently supported.
|
|
31
|
-
|
|
32
|
-
A typescript version of this MCP client is available
|
|
33
|
-
[here](https://github.com/hideya/mcp-client-langchain-ts)
|
|
34
|
-
|
|
35
|
-
## Requirements
|
|
36
|
-
|
|
37
|
-
- Python 3.11+
|
|
38
|
-
- [`uv`](https://docs.astral.sh/uv/) installation
|
|
39
|
-
- API keys from [Anthropic](https://console.anthropic.com/settings/keys),
|
|
40
|
-
[OpenAI](https://platform.openai.com/api-keys), and/or
|
|
41
|
-
[Groq](https://console.groq.com/keys)
|
|
42
|
-
as needed
|
|
43
|
-
|
|
44
|
-
## Setup
|
|
45
|
-
1. Install dependencies:
|
|
46
|
-
```bash
|
|
47
|
-
make install
|
|
48
|
-
```
|
|
49
|
-
|
|
50
|
-
2. Setup API keys:
|
|
51
|
-
```bash
|
|
52
|
-
cp .env.template .env
|
|
53
|
-
```
|
|
54
|
-
- Update `.env` as needed.
|
|
55
|
-
- `.gitignore` is configured to ignore `.env`
|
|
56
|
-
to prevent accidental commits of the credentials.
|
|
57
|
-
|
|
58
|
-
3. Configure LLM and MCP Servers settings `llm_mcp_config.json5` as needed.
|
|
59
|
-
|
|
60
|
-
- [The configuration file format](https://github.com/hideya/mcp-client-langchain-ts/blob/main/llm_mcp_config.json5)
|
|
61
|
-
for MCP servers follows the same structure as
|
|
62
|
-
[Claude for Desktop](https://modelcontextprotocol.io/quickstart/user),
|
|
63
|
-
with one difference: the key name `mcpServers` has been changed
|
|
64
|
-
to `mcp_servers` to follow the snake_case convention
|
|
65
|
-
commonly used in JSON configuration files.
|
|
66
|
-
- The file format is [JSON5](https://json5.org/),
|
|
67
|
-
where comments and trailing commas are allowed.
|
|
68
|
-
- The format is further extended to replace `${...}` notations
|
|
69
|
-
with the values of corresponding environment variables.
|
|
70
|
-
- Keep all the credentials and private info in the `.env` file
|
|
71
|
-
and refer to them with `${...}` notation as needed.
|
|
72
|
-
|
|
73
|
-
|
|
74
|
-
## Usage
|
|
75
|
-
|
|
76
|
-
Run the app:
|
|
77
|
-
```bash
|
|
78
|
-
make start
|
|
79
|
-
```
|
|
80
|
-
|
|
81
|
-
Run in verbose mode:
|
|
82
|
-
```bash
|
|
83
|
-
make start-v
|
|
84
|
-
```
|
|
85
|
-
|
|
86
|
-
See commandline options:
|
|
87
|
-
```bash
|
|
88
|
-
make start-h
|
|
89
|
-
```
|
|
90
|
-
|
|
91
|
-
At the prompt, you can simply press Enter to use example queries that perform MCP server tool invocations.
|
|
92
|
-
|
|
93
|
-
Example queries can be configured in `llm_mcp_config.json5`
|
|
@@ -1,12 +0,0 @@
|
|
|
1
|
-
LICENSE
|
|
2
|
-
README.md
|
|
3
|
-
pyproject.toml
|
|
4
|
-
src/cli_chat.py
|
|
5
|
-
src/config_loader.py
|
|
6
|
-
src/langchain_mcp_tools.py
|
|
7
|
-
src/langchain_mcp_tools.egg-info/PKG-INFO
|
|
8
|
-
src/langchain_mcp_tools.egg-info/SOURCES.txt
|
|
9
|
-
src/langchain_mcp_tools.egg-info/dependency_links.txt
|
|
10
|
-
src/langchain_mcp_tools.egg-info/requires.txt
|
|
11
|
-
src/langchain_mcp_tools.egg-info/top_level.txt
|
|
12
|
-
tests/test_langchain_mcp_tools.py
|
|
File without changes
|
|
File without changes
|
|
File without changes
|
|
File without changes
|
|
File without changes
|