Self-hosted OpenAI/Anthropic-compatible gateway: give each AI agent run a dollar budget, enforced before the provider, with a payload-free per-run cost record.
Setup
Choose your client, copy the config, then fill in any required environment variables.
claude mcp add inferrail -- uvx inferrail~/.codex/config.toml
[mcp_servers.inferrail]
command = "uvx"
args = ["inferrail"]
# codex mcp add inferrail -- uvx inferrail~/.cursor/mcp.json
{
"mcpServers": {
"inferrail": {
"command": "uvx",
"args": [
"inferrail"
]
}
}
}.vscode/mcp.json
{
"servers": {
"inferrail": {
"type": "stdio",
"command": "uvx",
"args": [
"inferrail"
]
}
}
}~/.gemini/settings.json
{
"mcpServers": {
"inferrail": {
"command": "uvx",
"args": [
"inferrail"
]
}
}
}claude_desktop_config.json
{
"mcpServers": {
"inferrail": {
"command": "uvx",
"args": [
"inferrail"
]
}
}
}.trae/mcp.json
{
"mcpServers": {
"inferrail": {
"command": "uvx",
"args": [
"inferrail"
]
}
}
}