liminality
physea/liminalityLiminality turns hard questions into worked, reusable routes. Hand solve() anything non-trivial: it breaks the problem down, grounds each piece against real tools, and returns a scored decision frame or a synthesized answer instead of a one-off guess. Solved routes are saved and re-runnable, so repeat questions come back faster and cheaper. Grade any result with report_outcome and the routes improve from use. Docs and setup: https://physea.ai/mcp
Watch this dependency
Get before/after evidence when observed tools, schemas, authentication access, availability or the endpoint changes.
Inspect evidence and set up a watch Share public evidence report
Score: 20/100
Our probe could not complete the MCP handshake.
why this score
How to use
Add to your Claude Desktop / Cursor / Cline MCP config:
{
"mcpServers": {
"liminality": {
"url": "https://physea.ai/mcp",
"transport": "http"
}
}
}