Skip to content

Commit 34b366e

Browse files
slimslenderslacksslimslenderslacks
committed
Add wayfound static content
1 parent 59f9540 commit 34b366e

File tree

2 files changed

+204
-0
lines changed

2 files changed

+204
-0
lines changed

‎servers/wayfound/readme.md‎

Lines changed: 178 additions & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -0,0 +1,178 @@
1+
# Wayfound MCP Server
2+
3+
[See description](https://www.wayfound.ai/pages/wayfound-mcp)
4+
5+
## What is the Wayfound MCP Remote Server?
6+
7+
The Wayfound MCP remote server is a specialized Model Context Protocol server that provides AI agents with access to organizational data and tools. It enables agents to:
8+
9+
- List and interact with agents in your organization
10+
- Get details about agents including their role, goal, guidelines, etc.
11+
- Get performance analysis of agents including guideline violations, knowledge gaps, user ratings, sentiment and more.
12+
- Get agent improvement suggestions based on performance analysis and feedback.
13+
14+
MCP (Model Context Protocol) is an open standard that allows AI applications to connect to external data sources and tools in a secure, standardized way. Think of it as a universal connector that lets your AI agents interact with real-world services and data.
15+
16+
## Available MCP Tools
17+
18+
The Wayfound MCP server provides the following tools for interacting with your organization's agents:
19+
20+
### `list_agents`
21+
**Description**: Get the list of all Agents in your Wayfound organization.
22+
23+
**Usage**: Ask questions like "What agents are in my organization?" or "List all available agents"
24+
25+
**Returns**: A comprehensive list of all agents configured in your Wayfound organization, including their names and basic information.
26+
27+
### `get_agent_details`
28+
**Description**: Get the details of a specific Agent in your Wayfound organization.
29+
30+
**Usage**: Ask questions like "Tell me about the Customer Support agent" or "What are the details for agent X?"
31+
32+
**Returns**: Detailed information about a specific agent including:
33+
- Agent role and purpose
34+
- Goals and objectives
35+
- Guidelines and constraints
36+
- Configuration settings
37+
- Other relevant metadata
38+
39+
### `get_manager_analysis_for_agent`
40+
**Description**: Get Wayfound Manager analysis for a specific Agent. This includes top topics, potential issues, tool call data, knowledge gaps, user ratings, sentiment, and guideline issues.
41+
42+
**Usage**: Ask questions like "What's the performance analysis for my Sales agent?" or "Show me the manager analysis for agent X"
43+
44+
**Returns**: Comprehensive performance analysis including:
45+
- **Top Topics**: Most frequently discussed subjects
46+
- **Potential Issues**: Identified problems or concerns
47+
- **Tool Call Data**: Usage statistics and patterns
48+
- **Knowledge Gaps**: Areas where the agent lacks information
49+
- **User Ratings**: Customer satisfaction scores
50+
- **Sentiment Analysis**: Overall user sentiment trends
51+
- **Guideline Issues**: Violations or deviations from established guidelines
52+
53+
### `get_improvement_suggestions_for_agent`
54+
**Description**: Get improvement suggestions for a specific Agent. This includes suggested system prompt updates and additional knowledge needed.
55+
56+
**Usage**: Ask questions like "How can I improve my Customer Service agent?" or "What suggestions do you have for agent X?"
57+
58+
**Returns**: Actionable improvement recommendations including:
59+
- **System Prompt Updates**: Suggested modifications to the agent's instructions
60+
- **Additional Knowledge**: Recommended knowledge base additions
61+
- **Training Suggestions**: Areas for further development
62+
- **Best Practice Recommendations**: Industry-standard improvements
63+
64+
## Example Queries
65+
66+
Here are some example questions you can ask your agent:
67+
68+
```python
69+
# Basic agent information
70+
"What are all the agents in my organization?"
71+
"Give me details about the Customer Support agent"
72+
73+
# Performance analysis
74+
"Show me the manager analysis for my Sales agent"
75+
"What are the top topics and issues for the Marketing agent?"
76+
77+
# Improvement recommendations
78+
"How can I improve my Customer Service agent's performance?"
79+
"What knowledge gaps exist for the Technical Support agent?"
80+
81+
# Combined queries
82+
"List all agents and show me which ones have the most issues"
83+
"Compare the performance of my Sales and Marketing agents"
84+
```
85+
86+
87+
## Project Structure
88+
89+
```
90+
wayfound-mcp-example/
91+
├── main.py # Main example script
92+
├── .env # Environment variables (create this file based on .env.example)
93+
├── README.md # This file
94+
└── requirements.txt # Python dependencies
95+
```
96+
97+
## Setup Instructions
98+
99+
### 1. Clone and Navigate
100+
101+
```bash
102+
git clone https://github.com/Wayfound-AI/wayfound-mcp-example
103+
cd wayfound-mcp-example
104+
```
105+
106+
### 2. Install Dependencies
107+
108+
Make sure you have Python 3.10+ installed, then install the required packages:
109+
110+
```bash
111+
pip install -r requirements.txt
112+
```
113+
114+
### 3. Configure Environment Variables
115+
116+
Create a `.env` file in the project root with the following configuration:
117+
118+
```env
119+
# Wayfound MCP Server Configuration
120+
WAYFOUND_MCP_API_KEY=your_mcp_api_key_here
121+
122+
# OpenAI Configuration (required for the Agents SDK)
123+
OPENAI_API_KEY=your_openai_api_key_here
124+
```
125+
126+
**Important**:
127+
- Replace `your_api_key_here` with your actual Wayfound MCP API key
128+
- Replace `your_openai_api_key_here` with your OpenAI API key
129+
- Keep the `.env` file secure and never commit it to version control
130+
131+
## Running the Example
132+
133+
Run the main example script:
134+
135+
```bash
136+
python main.py
137+
```
138+
139+
The script will:
140+
141+
1. Connect to the Wayfound MCP remote server using SSE (Server-Sent Events)
142+
2. Create an AI agent with access to MCP tools
143+
3. Execute example queries to demonstrate the available functionality
144+
4. Display the results and provide a trace URL for debugging
145+
146+
### Example Output
147+
148+
```
149+
View trace: https://platform.openai.com/traces/trace?trace_id=<trace_id>
150+
151+
Running: What are all the agents in my organization?
152+
[Agent response with list of organizational agents]
153+
```
154+
155+
## Customizing the Example
156+
157+
You can modify `main.py` to:
158+
159+
- Change the agent instructions
160+
- Add new queries or tool calls
161+
- Adjust model settings
162+
- Add error handling and logging
163+
164+
## Troubleshooting
165+
166+
### Debug Mode
167+
168+
Enable debug logging to see more detailed information:
169+
170+
```python
171+
import logging
172+
logging.basicConfig(level=logging.DEBUG)
173+
```
174+
175+
## Learn More
176+
177+
- [OpenAI Agents SDK Documentation](https://openai.github.io/openai-agents-python/)
178+
- [Model Context Protocol Specification](https://modelcontextprotocol.io/)

‎servers/wayfound/tools.json‎

Lines changed: 26 additions & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -0,0 +1,26 @@
1+
[
2+
{
3+
"name": "list_agents",
4+
"description": "Get the list of all Agents in your Wayfound organization.",
5+
"arguments": [
6+
]
7+
},
8+
{
9+
"name": "get_agent_details",
10+
"description": "Get the details of a specific Agent in your Wayfound organization.",
11+
"arguments": [
12+
]
13+
},
14+
{
15+
"name": "get_manager_analysis_for_agent",
16+
"description": "Get the details of a specific Agent in your Wayfound organization.",
17+
"arguments": [
18+
]
19+
},
20+
{
21+
"name": "get_improvement_suggestions_for_agents",
22+
"description": "Get improvement suggestions for a specific Agent. This includes suggested system prompt updates and additional knowledge needed.",
23+
"arguments": [
24+
]
25+
}
26+
]

0 commit comments

Comments
 (0)