A community-driven registry for the Claude Code ecosystem. Not affiliated with Anthropic.
Are you the author? Sign in to claim
MCP server for persona-driven AI sessions: practice interviews, conduct guided reflection, and role-play conversations w
A Model Context Protocol (MCP) server that enables AI assistants to conduct structured, persona-driven sessions including interview preparation, personal reflection, and coaching conversations.
Transform your AI interactions with realistic persona-driven sessions. Originally designed for mock interview preparation (practice with your "new boss" before that important meeting), this MCP server has evolved into a flexible framework for guided conversations of all kinds.
Whether you're preparing for a crucial presentation, seeking structured self-reflection, or wanting to practice difficult conversations in a safe environment, this server provides the framework for meaningful, guided dialogue.
Practice conversations with realistic personas:
Adaptive journaling sessions that flow between:
Create your own personas and session frameworks for specific needs.
git clone https://github.com/mattjoyce/mcp-persona-sessions.git
cd mcp-persona-sessions
pip install fastmcp
cp config.yaml.example config.yaml
# Edit config.yaml with your preferences
Add to your MCP client configuration (e.g., Claude Desktop):
{
"mcpServers": {
"persona-sessions": {
"command": "/path/to/venv/python",
"args": ["/path/to/mcp-persona-sessions/mcp-persona-sessions.py"],
"cwd": "/path/to/mcp-persona-sessions"
}
}
}
I use Claude as the mcp client, you might use something else. Simple ask claude to initiate a session.
"Claude, use your tools to run a mock meeting, this meeting is with XXX and is about YYY, duration 10 mins." If you have useful documents, include them with the request.
persona_path: "roles" # Directory containing persona files
session_types_file: "session_types.yaml" # Session framework definitions
default_persona_file: "Role-Interviewer-mcp.md" # Fallback persona
evaluator_persona_file: "Role-Interview-Evaluator.md" # Feedback persona
Create a new .md file in the roles/ directory with:
# Your Persona Name
## Identity and Purpose
[Define who this persona is and their role]
## Communication Style
[How they speak and interact]
## Expertise Areas
[What they know and focus on]
## Goals
[What they aim to achieve in sessions]
[Additional sections as needed...]
Edit session_types.yaml to add new session frameworks:
session_types:
- name: "Your Custom Session"
persona: "Your Persona"
persona_file: "Role-Your-Persona.md"
description: "What this session accomplishes"
prerequisites: ["What's needed to start"]
duration: 20
topics: ["Key areas to explore"]
goals: ["Session outcomes"]
mcp-persona-sessions/
├── mcp-persona-sessions.py # Main MCP server
├── timer.py # Timer management utilities
├── config.yaml # Server configuration
├── session_types.yaml # Available session frameworks
├── roles/ # Persona definition files
│ ├── Role-CIO.md
│ ├── Role-Reflective-Companion.md
│ ├── Role-Interview-Evaluator.md
│ └── ...
├── requirements.txt
└── README.md
| Tool | Purpose |
|---|---|
list_session_frameworks | Browse available session types and templates |
list_session_personas | View all available personas |
get_persona_details | Get full details of a specific persona |
assess_session_readiness | Check if ready to start a session |
initialize_session | Begin a structured session |
start_timer | Start session timer with optional duration |
check_timer_status | Monitor running timer progress |
stop_timer | End timer and conclude session |
get_session_feedback | Analyze session transcript for feedback |
Contributions welcome! Areas of interest:
GPL-3.0 License - see LICENSE file for details.
Built using the Model Context Protocol framework for AI assistant extensibility.
See CHANGELOG.md.
Run Claude Code as an MCP server so any agent can delegate coding tasks to it
Browser automation using accessibility snapshots instead of screenshots
MCP server integration for DaVinci Resolve Studio
Secure MCP server for MySQL database interaction, queries, and schema management