tokencast runs on your own machine — the client starts it, so there is no endpoint to ping. 57 installs a week from pypi. Last commit 8 Apr 2026.
Pre-execution cost estimation for LLM agent workflows with calibration learning
This server runs on your own machine — install it with the package manager and the client starts it for you. Package name taken from the official registry entry.
claude mcp add tokencast -- uvx tokencast
{
"mcpServers": {
"tokencast": {
"args": [
"tokencast"
],
"command": "uvx"
}
}
}
[mcp_servers.tokencast]
command = "uvx"
args = ["tokencast"]
{
"mcpServers": {
"tokencast": {
"args": [
"tokencast"
],
"command": "uvx"
}
}
}
{
"mcpServers": {
"tokencast": {
"args": [
"tokencast"
],
"command": "uvx"
}
}
}
Answers built from our own checks of this server.