-
Notifications
You must be signed in to change notification settings - Fork 36
Open
Description
Describe your feature request
I would like to be able to pull in read-only toolsets that are available from the Render MCP server.
This would allow the local LLM models to 👉 safely 👈 call tools which do not make any mutations / changes to Render platform / databases, and rather performs retrieval /queries of structured data from the Render platform, thus enabling fine-grained access-controls, avoiding potential unsafe LLM-driven accidental data writes to the Render platform.
A similar configuration can be considered like we see with the GitHub MCP server configuration 👇
{
"servers": {
"github": {
"type": "http",
"url": "https://api.githubcopilot.com/mcp/",
"headers": {
"X-MCP-Readonly": "true" // 👈 to pull down and access limited read-only toolsets
}
}
}
}Render MCP Config - Similar Suggestion?
{
"mcpServers": {
"render": {
"url": "https://mcp.render.com/mcp",
"headers": {
"Authorization": "Bearer <YOUR_API_KEY>"
"X-MCP-Readonly": "true" // 👈 to pull down and access limited read-only toolsets
}
}
}
}Metadata
Metadata
Assignees
Labels
No labels