lastseen score
Trust scores and dated liveness/drift facts for MCP servers, from the lastseen.dev notary register.
Should I use this
Quality & Safety
Based on automated analysis of tool definitions and protocol compliance.
Context Cost
This is the approximate number of tokens consumed each time the server's tools are loaded into a model's context. Higher counts reduce the attention available for other tasks.
Install
One-Click Install
Add this to your `claude_desktop_config.json` file:
{
"mcpServers": {
"score": {
"url": "https://score.lastseen.dev/mcp"
}
}
}Remote endpoints
https://score.lastseen.dev/mcpstreamable-httpWhat it can do
Tool inventory
Tools (3)
⚪trust_score(id)
Derived trust score (0-100) for an MCP server, computed from the lastseen.dev register's dated observations by a published method. The score carries DEPTH (how many of 5 dimensions were computable) BOUND INTO the number as `display`, e.g. '80 · depth 2/5 · provisional' — a depth-2 score answers a narrower question (mortality, not merit). Not a prediction, not advice. Methodology: score.lastseen.dev/methodology.
Input Schema
{
"type": "object",
"properties": {
"id": {
"type": "string",
"description": "MCP schema-server identifier, e.g. io.github.owner/name"
}
},
"required": [
"id"
]
}🟢schema_history(id)
Dated schema/tool-count observations for an MCP server (a series when known): first_seen, last_observed, drift events. Facts from the register, not a score.
Input Schema
{
"type": "object",
"properties": {
"id": {
"type": "string",
"description": "e.g. io.github.owner/name"
}
},
"required": [
"id"
]
}⚪dependency_risk(id)
An MCP server's dependencies and their mortality state (dead/dormant counts) from the register. Facts, not a score.
Input Schema
{
"type": "object",
"properties": {
"id": {
"type": "string",
"description": "e.g. io.github.owner/name"
}
},
"required": [
"id"
]
}Community
Evidence