io.github.ribato22/volanteother

Volante

Verified · 36 days ago

Routes to the best predicted fit from configured metadata/evidence after hard-capability filtering.

Install

claude mcp add volante -- uvx volante

Requires: ANTHROPIC_API_KEY, ANTHROPIC_MODELS, ANTHROPIC_NAMES, ANTHROPIC_STRENGTHS, ANTHROPIC_CONTEXT, ANTHROPIC_MAX_OUTPUT, ANTHROPIC_TOOLS, ANTHROPIC_COST_IN, ANTHROPIC_COST_OUT, ANTHROPIC_TIER, MOONSHOT_API_KEY, MOONSHOT_BASE_URL, MOONSHOT_MODELS, MOONSHOT_NAMES, MOONSHOT_STRENGTHS, MOONSHOT_CONTEXT, MOONSHOT_MAX_OUTPUT, MOONSHOT_TOOLS, MOONSHOT_COST_IN, MOONSHOT_COST_OUT, MOONSHOT_TIER, OLLAMA_BASE_URL, OLLAMA_API_KEY, OLLAMA_MODELS, OLLAMA_NAMES, OLLAMA_STRENGTHS, OLLAMA_CONTEXT, OLLAMA_MAX_OUTPUT, OLLAMA_TOOLS, OLLAMA_COST_IN, OLLAMA_COST_OUT, OLLAMA_TIER, CLAUDE_CODE_ENABLED, CLAUDE_CODE_MODELS, CLAUDE_CODE_TIER, CLAUDE_CODE_MAX_OUTPUT, CLAUDE_CODE_TIMEOUT, CLAUDE_CODE_SYSTEM_PROMPT_MODE, CODEX_ENABLED, CODEX_MODELS, CODEX_TIER, CODEX_CONTEXT, CODEX_MAX_OUTPUT, CODEX_COST_IN, CODEX_COST_OUT, CODEX_TIMEOUT, OPENAI_COMPAT_BASE_URL, OPENAI_COMPAT_KEY, OPENAI_COMPAT_MODEL, OPENAI_COMPAT_NAME, OPENAI_COMPAT_STRENGTHS, OPENAI_COMPAT_CONTEXT, OPENAI_COMPAT_MAX_OUTPUT, OPENAI_COMPAT_TOOLS, OPENAI_COMPAT_COST_IN, OPENAI_COMPAT_COST_OUT, OPENAI_COMPAT_TIER, OPENAI_COMPAT_2_BASE_URL, OPENAI_COMPAT_2_KEY, OPENAI_COMPAT_2_MODEL, OPENAI_COMPAT_2_NAME, OPENAI_COMPAT_2_STRENGTHS, OPENAI_COMPAT_2_CONTEXT, OPENAI_COMPAT_2_MAX_OUTPUT, OPENAI_COMPAT_2_TOOLS, OPENAI_COMPAT_2_COST_IN, OPENAI_COMPAT_2_COST_OUT, OPENAI_COMPAT_2_TIER, OPENAI_COMPAT_3_BASE_URL, OPENAI_COMPAT_3_KEY, OPENAI_COMPAT_3_MODEL, OPENAI_COMPAT_3_NAME, OPENAI_COMPAT_3_STRENGTHS, OPENAI_COMPAT_3_CONTEXT, OPENAI_COMPAT_3_MAX_OUTPUT, OPENAI_COMPAT_3_TOOLS, OPENAI_COMPAT_3_COST_IN, OPENAI_COMPAT_3_COST_OUT, OPENAI_COMPAT_3_TIER, VOLANTE_QUALITY_PROFILES_FILE, VOLANTE_MODEL_OVERRIDES_FILE, VOLANTE_FETCH_ALLOWLIST, VOLANTE_READ_ROOT, VOLANTE_SANDBOX, VOLANTE_MAX_SUBSCRIPTION_CALLS

Our take

This server seems to route requests to the best predicted tool or target based on configured metadata and evidence after applying hard capability filters. It is for developers building orchestration or tool-selection layers inside larger MCP-based systems. The concept is reasonable, but the description is abstract and the listing shows no adoption signal, so it currently reads as an experimental infrastructure component.

reviewed by hand · 2026-07-28

Something wrong or dead here? Report it