A side-by-side look at the Openrouter and Promptheus MCP servers: transport, auth, tools, quality signals, and the config to install either one in Claude, Cursor, Windsurf or VS Code.
Verdict: choose Openrouter for conversational ai work as a local process; choose Promptheus for conversational ai as a local process. You can also run both side by side in one mcpServers block.
list_models, get_model, compare_models.GOOGLE_API_KEY, ANTHROPIC_API_KEY, OPENAI_API_KEY, GROQ_API_KEY, DASHSCOPE_API_KEY, ZHIPUAI_API_KEY, OPENROUTER_API_KEY.refine_prompt, tweak_prompt, list_models.Rows where the two servers differ are marked.
| Spec | ||
|---|---|---|
| Category | Conversational AI | Conversational AI |
| Quality signal | 51/100 · Good | 55/100 · Good |
| Install command | Confirmed | Confirmed |
| Transport | Local subprocess (stdio) | Local subprocess (stdio) |
| Auth | Not declared | Not declared |
| Pricing | Not declared | Not declared |
| Required env vars | None listed | GOOGLE_API_KEYANTHROPIC_API_KEYOPENAI_API_KEYGROQ_API_KEYDASHSCOPE_API_KEYZHIPUAI_API_KEYOPENROUTER_API_KEY |
| Verified owner | Community listing | Community listing |
| GitHub stars | 0 | 17 |
| Directory activity | 0 views 0 installs 0 upvotes | 0 views 0 installs 0 upvotes |
| Tools | 3 | 5 |
| Full listing | View Openrouter | View Promptheus |
list_modelsList LLMs from the OpenRouter catalog (~340 models across all major providers) with normalized $/1M-token pricing, context windows, and modality. Filter by search, free-only, minimum context, or modality (e.g. "image" for vision models). Sorted cheapest-first by completion price. Keyless.get_modelGet full details for one OpenRouter model by id — normalized $/1M pricing (prompt, completion, plus per-request/image/web-search costs), context window, modality, supported parameters, tokenizer, knowledge cutoff. e.g. id "anthropic/claude-sonnet-4.5". Keyless.compare_modelsSide-by-side comparison of pricing ($/1M tokens), context window, and modality for 2-5 OpenRouter models. Answers "is claude-sonnet-4.5 cheaper than gpt-4o". Keyless.refine_promptIntelligent prompt refinement with optional clarification questions.tweak_promptApply targeted modifications to existing prompts.list_modelsDiscover available models from configured providers.list_providersCheck provider configuration status.validate_environmentTest environment configuration and API connectivity.Paste either block (or both) into your client config, e.g. claude_desktop_config.json or ~/.cursor/mcp.json. We only show a config when the install command is confirmed.
{
"mcpServers": {
"openrouter": {
"command": "npx",
"args": [
"-y",
"@pipeworx/mcp-openrouter"
]
}
}
}
{
"mcpServers": {
"promptheus": {
"command": "uvx",
"args": [
"promptheus"
],
"env": {
"GOOGLE_API_KEY": "YOUR_GOOGLE_API_KEY_HERE",
"ANTHROPIC_API_KEY": "YOUR_ANTHROPIC_API_KEY_HERE",
"OPENAI_API_KEY": "YOUR_OPENAI_API_KEY_HERE",
"GROQ_API_KEY": "YOUR_GROQ_API_KEY_HERE",
"DASHSCOPE_API_KEY": "YOUR_DASHSCOPE_API_KEY_HERE",
"ZHIPUAI_API_KEY": "YOUR_ZHIPUAI_API_KEY_HERE",
"OPENROUTER_API_KEY": "YOUR_OPENROUTER_API_KEY_HERE"
}
}
}
}