MCP server that completes prompts with automatic provider fallback across OpenAI, Anthropic Claude, and Google Gemini. Handles rate limits, transient errors, and missing keys by trying the next provider in chain. Returns first success with full attempt lo
- Runs on your machine
- No known advisories
- MIT
Version checks
- Available on npm, latest 0.1.0 · checked
Not run yet – package checks are being switched on. It runs from a package on your machine, so instead of uptime we will install and start each new version.
Run it on your machine
Add to claude_desktop_config.json (Settings › Developer › Edit Config):
Add to ~/.cursor/mcp.json:
Add to ~/.config/devin/mcp_config.json:
Add to cline_mcp_settings.json (MCP Servers › Configure › Configure MCP Servers):
Add to mcp.json (Program › Install › Edit mcp.json):
Add to ~/.gemini/settings.json:
Add to ~/.llm/config.json:
Add to config.json in oterm's data folder (~/Library/Application Support/oterm on macOS, ~/.local/share/oterm on Linux, %APPDATA%\oterm on Windows):
Add to resources/assets/config/mcp.json in the installed app:
Add to ~/.boltai/mcp.json:
Add to anythingllm_mcp_servers.json in the AnythingLLM storage plugins directory:
Add to Augment Settings Panel – MCP section – Import from JSON (VS Code extension):
Add to ~/.bob/settings/mcp.json (global) or .bob/mcp.json (project):
Add to ~/.config/eca/config.json (global) or .eca/config.json in the project:
Add to ~/.kiro/settings/mcp.json (user) or .kiro/settings/mcp.json (workspace):
Add to ~/.kiro/settings/mcp.json (user) or .kiro/settings/mcp.json (workspace):
{
"mcpServers": {
"llm-fallback-mcp": {
"command": "npx",
"args": ["-y", "llm-fallback-mcp"]
}
}
} Step by step for Claude Desktop
Step by step for Devin Desktop
Run in a terminal:
claude mcp add --transport stdio llm-fallback-mcp -- npx -y llm-fallback-mcp Add to .vscode/mcp.json:
{
"servers": {
"llm-fallback-mcp": {
"type": "stdio",
"command": "npx",
"args": ["-y", "llm-fallback-mcp"]
}
}
} Add to ~/.config/zed/settings.json:
{
"context_servers": {
"llm-fallback-mcp": {
"command": "npx",
"args": ["-y", "llm-fallback-mcp"]
}
}
} Add to ~/.continue/config.yaml:
mcpServers:
- name: "llm-fallback-mcp"
type: stdio
command: "npx"
args:
- "-y"
- "llm-fallback-mcp" Add to ~/.config/goose/config.yaml:
extensions:
"llm-fallback-mcp":
type: stdio
name: "llm-fallback-mcp"
enabled: true
cmd: "npx"
args: ["-y", "llm-fallback-mcp"]
env_keys: []
envs: {}
timeout: 300 Add to ~/.config/agenvoy/mcp.json:
{
"servers": {
"llm-fallback-mcp": {
"command": "npx",
"args": ["-y", "llm-fallback-mcp"]
}
}
} Add to ~/.config/amp/settings.json (user) or .amp/settings.json (workspace):
{
"amp.mcpServers": {
"llm-fallback-mcp": {
"command": "npx",
"args": ["-y", "llm-fallback-mcp"]
}
}
} For agents: the install prompt
Every step above as one Markdown text, for an assistant installing it for you.
# Install the MCP server llm-fallback-mcp (`io.github.khushalb25/llm-fallback-mcp`)
Its page: https://plug.directory/servers/io.github.khushalb25/llm-fallback-mcp
It runs on the person's machine and needs Node.js installed.
Use the section for the person's client.
## Claude Desktop, Cursor, Devin Desktop, Cline, LM Studio, Gemini CLI, mcp-client-cli, oterm, TUUI, BoltAI, AnythingLLM, Augment Code, Bob Shell, ECA, Kiro, Kiro CLI
Merge into the place for the person's client:
- Claude Desktop: claude_desktop_config.json (Settings › Developer › Edit Config)
- Cursor: ~/.cursor/mcp.json
- Devin Desktop: ~/.config/devin/mcp_config.json
- Cline: cline_mcp_settings.json (MCP Servers › Configure › Configure MCP Servers)
- LM Studio: mcp.json (Program › Install › Edit mcp.json)
- Gemini CLI: ~/.gemini/settings.json
- mcp-client-cli: ~/.llm/config.json
- oterm: config.json in oterm's data folder (~/Library/Application Support/oterm on macOS, ~/.local/share/oterm on Linux, %APPDATA%\oterm on Windows)
- TUUI: resources/assets/config/mcp.json in the installed app
- BoltAI: ~/.boltai/mcp.json
- AnythingLLM: anythingllm_mcp_servers.json in the AnythingLLM storage plugins directory
- Augment Code: Augment Settings Panel – MCP section – Import from JSON (VS Code extension)
- Bob Shell: ~/.bob/settings/mcp.json (global) or .bob/mcp.json (project)
- ECA: ~/.config/eca/config.json (global) or .eca/config.json in the project
- Kiro: ~/.kiro/settings/mcp.json (user) or .kiro/settings/mcp.json (workspace)
- Kiro CLI: ~/.kiro/settings/mcp.json (user) or .kiro/settings/mcp.json (workspace)
```json
{
"mcpServers": {
"llm-fallback-mcp": {
"command": "npx",
"args": ["-y", "llm-fallback-mcp"]
}
}
}
```
## Claude Code
Run in a terminal:
```sh
claude mcp add --transport stdio llm-fallback-mcp -- npx -y llm-fallback-mcp
```
## VS Code
Merge into .vscode/mcp.json:
```json
{
"servers": {
"llm-fallback-mcp": {
"type": "stdio",
"command": "npx",
"args": ["-y", "llm-fallback-mcp"]
}
}
}
```
## Zed
Merge into ~/.config/zed/settings.json:
```json
{
"context_servers": {
"llm-fallback-mcp": {
"command": "npx",
"args": ["-y", "llm-fallback-mcp"]
}
}
}
```
## Continue
Merge into ~/.continue/config.yaml:
```yaml
mcpServers:
- name: "llm-fallback-mcp"
type: stdio
command: "npx"
args:
- "-y"
- "llm-fallback-mcp"
```
## Goose
Merge into ~/.config/goose/config.yaml:
```yaml
extensions:
"llm-fallback-mcp":
type: stdio
name: "llm-fallback-mcp"
enabled: true
cmd: "npx"
args: ["-y", "llm-fallback-mcp"]
env_keys: []
envs: {}
timeout: 300
```
## Codex
Run in a terminal:
```sh
codex mcp add llm-fallback-mcp -- npx -y llm-fallback-mcp
```
## Agenvoy
Merge into ~/.config/agenvoy/mcp.json:
```json
{
"servers": {
"llm-fallback-mcp": {
"command": "npx",
"args": ["-y", "llm-fallback-mcp"]
}
}
}
```
## Amp
Merge into ~/.config/amp/settings.json (user) or .amp/settings.json (workspace):
```json
{
"amp.mcpServers": {
"llm-fallback-mcp": {
"command": "npx",
"args": ["-y", "llm-fallback-mcp"]
}
}
}
```
Then restart the client, or reload its MCP servers, and ask it to list the server's tools to confirm it started.
Known advisories
None known for its packages.
Checked 2026-10-28 against OSV.dev and deps.dev.
From npm. We have not checked this server yet. Report a problem