code firewall mcp for VS Code
io.github.egoughnour/code-firewall-mcp
Structural similarity-based code filter. Stops malicious code pattern reaching execution tools.
client:VS Code
transport:stdio
runtime:pypi
Install code firewall mcp in VS Code
.vscode/mcp.json
{
"inputs": [
{
"type": "promptString",
"id": "firewall-data-dir",
"description": "FIREWALL_DATA_DIR",
"password": true
},
{
"type": "promptString",
"id": "ollama-url",
"description": "OLLAMA_URL",
"password": true
},
{
"type": "promptString",
"id": "embedding-model",
"description": "EMBEDDING_MODEL",
"password": true
},
{
"type": "promptString",
"id": "similarity-threshold",
"description": "SIMILARITY_THRESHOLD",
"password": true
}
],
"servers": {
"code-firewall-mcp": {
"type": "stdio",
"command": "uvx",
"args": [
"code-firewall-mcp"
],
"env": {
"FIREWALL_DATA_DIR": "${input:firewall-data-dir}",
"OLLAMA_URL": "${input:ollama-url}",
"EMBEDDING_MODEL": "${input:embedding-model}",
"SIMILARITY_THRESHOLD": "${input:similarity-threshold}"
}
}
}
}
VS Code prompts for each input the first time the server starts, so no secret is written to the file. VS Code docs