Metal Mantra
io.metalmantrav1.0.0Updated Oct 9, 2026
Read-only evidence reports on AI agents: search, fetch a report, check trust. Not a certification.
Overview
AI-generated overview
A remote, read-only MCP endpoint for searching and fetching evidence reports about AI agents and checking their trust status.
- What it does
- Metal Mantra exposes a remote streamable HTTP endpoint that lets an assistant search for evidence reports on AI agents, fetch an individual report, and check a trust indicator. The description frames the output as read-only evidence reports and explicitly states it is not a certification. No tool list is published in the manifest, so the exact tool names and parameters are not documented here.
- When to use it
- Worth adding when an assistant needs background evidence about a specific AI agent or vendor and a quick trust check, for example during evaluation or due diligence. It is not a certification service, so it should not be treated as an authoritative approval or compliance verdict.
- Requirements
- A remote MCP client that supports streamable HTTP, pointed at the provider's endpoint. The manifest declares no authentication, no environment variables, and no headers, and no local package or runtime is needed. Network access to the provider is required.
Before you install
The manifest declares no authentication, so requests are not tied to a user account and the provider can see what is queried. Reports are third-party evidence, not certification, so verify important findings independently before relying on them. No write, payment, or credential-handling behavior is described.
Installation
In SourceWeft
- Open Metal Mantra in the dashboard and add it to a workspace.
- Enable the server for the chats that should use its tools.
Web executable via Streamable HTTP. Remote servers run from the web runtime once configured in a workspace.
Other MCP clients
Add this to your client's mcpServers config.
{
"mcpServers": {
"registry": {
"type": "http",
"url": "https://metalmantra.io/mcp"
}
}
}Tools
0Tool metadata has not been indexed yet.
Version history
1- v1.0.0LatestOct 9, 2026

