Nexus MCP is an MCP server that enables AI models to invoke AI CLI agents in parallel, providing features like automatic retries, JSON-first response parsing, and persistent preferences.
From the registry: Japanese LLM security — prompt injection detection (jpi-guard) + PII masking (PII Guard). Free.
Please install the `nexus-mcp` MCP server into my current AI client (that's you).
Required prerequisites (do these first if not already done):
- **uv / uvx** — Install uv (https://docs.astral.sh/uv/)
Optional prerequisites:
- AI CLI binaries (optional) — Gemini CLI, Codex, Claude Code, or OpenCode — required only for runners you want to use
Canonical MCP server config (stdio transport):
- command: `uvx`
- args: ["nexus-mcp"]
- optional environment variables:
- `NEXUS_GEMINI_MODEL`: Default Gemini model (example: `gemini-3-flash-preview`)
- `NEXUS_CODEX_MODEL`: Default Codex model (example: `gpt-5.2`)
- `NEXUS_OPENCODE_MODEL`: Default OpenCode model (example: `ollama-cloud/kimi-k2.5`)
Note: Invokes AI CLI agents (Gemini, Codex, Claude Code, OpenCode) as MCP tools with parallel execution and retries.
Add this MCP server to my current client's config in the correct format for you. If you need secrets or credentials I haven't provided, ASK me — do not invent values or leave raw placeholders. After adding it, tell me how to verify the server is connected.NEXUS_GEMINI_MODELrequiredSpecifies the model to use for Gemini CLI.NEXUS_CODEX_MODELrequiredSpecifies the model to use for Codex.NEXUS_OPENCODE_MODELrequiredSpecifies the model to use for OpenCode.AI orchestration with hive-mind swarms, neural networks, and 87 MCP tools for enterprise dev.
Persistent codebase knowledge graph. Survives session restarts and context compaction.
A powerful toolkit for coding, providing semantic retrieval and editing capabilities.