Model Context Protocol (MCP)
Fourth example in AI agents series. Here, I build a customer MCP server to give any AI app access to a toolset for an Artificial Virtual Assistant (AVA).
Links
How to run this example
- Clone this repo
- Install uv if you haven't already
# Mac/Linux
curl -LsSf https://astral.sh/uv/install.sh | sh
# Windows
powershell -ExecutionPolicy ByPass -c "irm https://astral.sh/uv/install.ps1 | iex"
- Test the server in dev mode
uv run mcp dev mcp-server-example.py
- Add server config to AI app (e.g. Claude Desktop or Cursor).
{
"mcpServers": {
"AVA": {
"command": "/Users/shawhin/.local/bin/uv", # replace with global path to your uv installation
"args": [
"--directory",
"/Users/shawhin/Documents/_code/_stv/sandbox/ava-mcp/", # replace with global path to repo
"run",
"mcp-server-example.py"
]
}
}
}
Customizing AVA's Behavior
Update Personal Details and Preferences
- Locate the
prompts/ava.mdfile in your project directory - Customize the file with:
- Communication preferences
- Specific instructions for handling tasks
- Any other relevant guidelines for AVA
Environment Setup (.env)
- Create a
.envfile in the root directory of the project with the following variables:
USER_EMAIL=your_email_address
# Google OAuth Credentials
GOOGLE_CREDENTIALS_PATH=.config/ava-agent/credentials.json
GOOGLE_TOKEN_PATH=.config/ava-agent/token.json
Required Environment Variables:
USER_EMAIL: The Gmail address you want to use for this applicationGOOGLE_CREDENTIALS_PATH: Path to your Google OAuth credentials fileGOOGLE_TOKEN_PATH: Path where the Google OAuth token will be stored
Google OAuth Setup
1. Create Project Directory Structure
First, create the required directory structure:
mkdir -p .config/ava-agent
2. Set Up Google Cloud Project
- Go to the Google Cloud Console
- Create a new project or select an existing one
- Enable the Gmail API:
- In the navigation menu, go to "APIs & Services" > "Library"
- Search for "Gmail API"
- Click "Enable"
3. Create OAuth Credentials
-
In the Google Cloud Console:
- Go to "APIs & Services" > "Credentials"
- Click "Create Credentials" > "OAuth client ID"
- Choose "Desktop application" as the application type
- Give it a name (e.g., "AVA Gmail Client")
- Click "Create"
-
Download the credentials:
- After creation, click "Download JSON"
- Save the downloaded file as
credentials.jsonin.config/ava-agent/ - The file should contain your client ID and client secret
4. Configure OAuth Consent Screen
- In the Google Cloud Console:
- Go to "APIs & Services" > "OAuth consent screen"
- Choose "External" user type
- Fill in the required information:
- App name
- User support email
- Developer contact information
- Add the Gmail API scope:
https://www.googleapis.com/auth/gmail.modify - Add your email as a test user
- Complete the configuration
Signing into Google
Before the server can access you Gmail account you will need to authorize it. You can do this by running uv run oauth.py which does the following.
- Check for the presence of
token.json - If not found, it will initiate the Google OAuth authentication flow
- Guide you through the authentication process in your browser:
- You'll be asked to sign in to your Google account
- Grant the requested permissions
- The application will automatically save the token
- Generate and store the token automatically
Security Notes
File Protection
- Never commit your
.envfile ortoken.jsonto version control - Keep your Google credentials secure
- Add the following to your
.gitignore:.env .config/ava-agent/token.json .config/ava-agent/credentials.json
Metadata
Release files for iflow-mcp_agent-mcp-uv 0.1.0
For a detailed explanation of source distributions (sdists) and built distributions (wheels), please see the package formats documentation.
Source distribution (sdist)
| File | Size | Uploaded | Provenance | Actions |
|---|---|---|---|---|
| iflow_mcp_agent_mcp_uv-0.1.0.tar.gz | 3.2 kB | Without provenance | Details for iflow_mcp_agent_mcp_uv-0.1.0.tar.gz |
Built distribution (wheel)
| File | Interpreter | ABI | Platform | Actions Reset |
|---|---|---|---|---|
| iflow_mcp_agent_mcp_uv-0.1.0-py3-none-any.whl | Python 3 | none | any | Details for iflow_mcp_agent_mcp_uv-0.1.0-py3-none-any.whl |
Total release size: 6.6 kB
Release files / iflow_mcp_agent_mcp_uv-0.1.0.tar.gz
| Download URL | iflow_mcp_agent_mcp_uv-0.1.0.tar.gz |
|---|---|
| Size | 3.2 kB |
| Tags | Source |
|
SHA-256 checksum How to use checksums |
05da16d33f5d3d26532f2cf4f3995ad63f613e9939ee4e6e3d8fb4b85452a6c2
|
|
BLAKE2b-256 checksum How to use checksums |
81b2ff1079cc8bb66155cbb1fef6352df8e3b7110aaa775a3c0246093851cf94
|
| Upload date | |
|
Uploaded using Trusted Publishing? What is trusted publishing? |
No |
| Uploaded via |
uv/0.9.10 {"installer":{"name":"uv","version":"0.9.10"},"python":null,"implementation":{"name":null,"version":null},"distro":{"name":"macOS","version":null,"id":null,"libc":null},"system":{"name":null,"release":null},"cpu":null,"openssl_version":null,"setuptools_version":null,"rustc_version":null,"ci":null}
|
Release files / iflow_mcp_agent_mcp_uv-0.1.0-py3-none-any.whl
| Download URL | iflow_mcp_agent_mcp_uv-0.1.0-py3-none-any.whl |
|---|---|
| Size | 3.4 kB |
| Tags | Python 3 |
|
SHA-256 checksum How to use checksums |
aa786644864a4882372a1cd3965067327b805e664e75ab1c14d1cc6cedf8527a
|
|
BLAKE2b-256 checksum How to use checksums |
9a093ab54659a27ecbb179e29679781eb6fa709313da54e1ad3f630e0fd53c1f
|
| Upload date | |
|
Uploaded using Trusted Publishing? What is trusted publishing? |
No |
| Uploaded via |
uv/0.9.10 {"installer":{"name":"uv","version":"0.9.10"},"python":null,"implementation":{"name":null,"version":null},"distro":{"name":"macOS","version":null,"id":null,"libc":null},"system":{"name":null,"release":null},"cpu":null,"openssl_version":null,"setuptools_version":null,"rustc_version":null,"ci":null}
|
Security for iflow-mcp_agent-mcp-uv 0.1.0
Provenance
Provenance describes where a file came from. On PyPI, provenance is shared via attestations, which provide a verifiable record of the build or publishing details. View details, limitations and caveats.
No provenance found
None of the files in this release were uploaded with provenance metadata.
A note on package security
PyPI is a repository for distributing packages, not a service for auditing them. While third-party security researchers and our staff do their best to detect and remove malicious packages, PyPI cannot guarantee the quality, security or suitability of any given package.
To learn more about protecting yourself from malicious actors, visit our package security documentation.