{"_id":"@iflow-mcp/osanoai-multicli","name":"@iflow-mcp/osanoai-multicli","dist-tags":{"latest":"1.5.28"},"versions":{"1.5.28":{"name":"@iflow-mcp/osanoai-multicli","version":"1.5.28","description":"MCP server for Multiple CLI coding agent integration (Claude + Gemini + Codex)","type":"module","main":"dist/index.js","bin":{"iflow-mcp_osanoai-multicli":"dist/index.js"},"scripts":{"build":"tsc && cp -f src/modelCatalog.generated.json dist/ 2>/dev/null || true","start":"node dist/index.js","dev":"tsc && node dist/index.js","test":"vitest run","test:watch":"vitest","test:coverage":"vitest run --coverage","lint":"tsc --noEmit","prepublishOnly":"echo '⚠️  Remember to test locally first!' && npm run build","refresh-catalog":"npx tsx scripts/refresh-catalog.ts","prepare":"husky || true"},"keywords":["mcp","modelcontextprotocol","gemini","codex","claude","cli","coding","agent","llm"],"author":{"name":"arlogilbert"},"repository":{"type":"git","url":"git+https://github.com/osanoai/multicli.git"},"bugs":{"url":"https://github.com/osanoai/multicli/issues"},"homepage":"https://github.com/osanoai/multicli#readme","engines":{"node":">=20.0.0"},"dependencies":{"@modelcontextprotocol/sdk":"^1.27.1","dotenv":"^17.3.1","zod":"^4.3.6"},"devDependencies":{"@types/node":"^25.3.3","@vitest/coverage-v8":"^4.0.18","husky":"^9.1.7","typescript":"^5.9.3","vitest":"^4.0.18"},"publishConfig":{"access":"public"},"types":"./dist/index.d.ts","_id":"@iflow-mcp/osanoai-multicli@1.5.28","_integrity":"sha512-kiizqDg1DN7xVN/1g2VwWzJuLEBppTSzaMvmWDwBuRpWCmJdzaxDIqw5g4/dgsumfUXH5X6ctid1Wa7jVTkEJQ==","_resolved":"/app/auto-mcp-upload/data/25243/iflow-mcp-osanoai-multicli-1.5.28.tgz","_from":"file:/app/auto-mcp-upload/data/25243/iflow-mcp-osanoai-multicli-1.5.28.tgz","_nodeVersion":"24.14.0","_npmVersion":"11.9.0","dist":{"integrity":"sha512-kiizqDg1DN7xVN/1g2VwWzJuLEBppTSzaMvmWDwBuRpWCmJdzaxDIqw5g4/dgsumfUXH5X6ctid1Wa7jVTkEJQ==","shasum":"1ac5aa2fdcedd4fefaf64e4a3f11cfac8d784b3b","tarball":"https://registry.npmjs.org/@iflow-mcp/osanoai-multicli/-/osanoai-multicli-1.5.28.tgz","fileCount":111,"unpackedSize":167040,"signatures":[{"keyid":"SHA256:DhQ8wR5APBvFHLF/+Tc+AYvPOdTpcIDqOhxsBHRwC7U","sig":"MEYCIQCh61kqKcauxuFK8HpN14A9KcbiM9WLAMCcYNeANwXmtgIhAJ/kGh6QG0YPpPeUl0hoSYynWmP13960vd85IkMVN3Ye"}]},"_npmUser":{"name":"chatflowdev","email":"chatflowdev@gmail.com"},"directories":{},"maintainers":[{"name":"chatflowdev","email":"chatflowdev@gmail.com"},{"name":"qystart","email":"987472953@qq.com"}],"_npmOperationalInternal":{"host":"s3://npm-registry-packages-npm-production","tmp":"tmp/osanoai-multicli_1.5.28_1773803158061_0.04127018444697894"},"_hasShrinkwrap":false}},"time":{"created":"2026-03-18T03:05:57.928Z","1.5.28":"2026-03-18T03:05:58.222Z","modified":"2026-03-18T03:05:58.587Z"},"maintainers":[{"name":"chatflowdev","email":"chatflowdev@gmail.com"},{"name":"qystart","email":"987472953@qq.com"}],"description":"MCP server for Multiple CLI coding agent integration (Claude + Gemini + Codex)","homepage":"https://github.com/osanoai/multicli#readme","keywords":["mcp","modelcontextprotocol","gemini","codex","claude","cli","coding","agent","llm"],"repository":{"type":"git","url":"git+https://github.com/osanoai/multicli.git"},"author":{"name":"arlogilbert"},"bugs":{"url":"https://github.com/osanoai/multicli/issues"},"readme":"# Multi-CLI MCP\n\n[![npm version](https://img.shields.io/npm/v/@osanoai/multicli)](https://www.npmjs.com/package/@osanoai/multicli)\n[![Tests](https://img.shields.io/github/actions/workflow/status/osanoai/multicli/tests.yml?branch=main&label=tests)](https://github.com/osanoai/multicli/actions/workflows/tests.yml)\n[![Scan](https://img.shields.io/github/actions/workflow/status/osanoai/multicli/scan.yml?branch=main&label=security%20scan)](https://github.com/osanoai/multicli/actions/workflows/scan.yml)\n[![GitHub release](https://img.shields.io/github/v/release/osanoai/multicli)](https://github.com/osanoai/multicli/releases/latest)\n[![Node](https://img.shields.io/node/v/@osanoai/multicli)](https://www.npmjs.com/package/@osanoai/multicli)\n[![TypeScript](https://img.shields.io/badge/TypeScript-5.x-3178c6?logo=typescript&logoColor=white)](https://www.typescriptlang.org/)\n\n**An MCP server that lets Claude, Gemini, Codex, and OpenCode call each other as tools.**\n\n```\nClaude:   \"Hey Gemini, what do you think about this code?\"\nGemini:   \"It's mass. Let me ask Codex for a second opinion.\"\nCodex:    \"You're both wrong. Here's the fix.\"\nOpenCode: \"I checked with three providers. They all agree with Codex.\"\n```\n\n---\n\n## One-Line Install\n\n```bash\ncurl -fsSL https://raw.githubusercontent.com/osanoai/multicli/main/install.sh | bash\n```\n\nDetects which AI CLIs you have installed and configures Multi-CLI for all of them automatically. No config files, no API keys, no environment variables. If it's on your PATH, it works.\n\n---\n\n## What It Does\n\nMulti-CLI sits between your AI clients and bridges them via the [Model Context Protocol](https://modelcontextprotocol.io/). Install it once, and whichever AI you're talking to gains the ability to call the others.\n\n- **Claude** can ask Gemini, Codex, or OpenCode for help\n- **Gemini** can delegate to Claude, Codex, or OpenCode\n- **Codex** can consult Claude, Gemini, or OpenCode\n- **OpenCode** can call Claude, Gemini, or Codex (across 75+ providers)\n- Each client's own tools are hidden (no talking to yourself, that's weird)\n- Auto-detects which CLIs you have installed — only shows what's available\n\n---\n\n## The Meta Part\n\nThis tool was built by the very AIs it connects.\n\nClaude, Gemini, Codex, and OpenCode wrote the code. Claude, Gemini, Codex, and OpenCode maintain it. Every night, a CI job queries the latest stable release of each CLI for its current model list, diffs the results against what's in the repo, and automatically publishes a new version if anything changed. New model releases get picked up within 24 hours. Deprecated models get cleaned out. The repo stays current without anyone touching it.\n\nBecause all install commands use `@latest`, your MCP client pulls the newest version every time it starts — no manual updates, no stale model lists, no maintenance.\n\nMost MCP tools go stale within weeks. This one is self-maintaining by design.\n\n---\n\n## Prerequisites\n\nYou need **Node.js >= 20** and at least **two** of these CLIs installed:\n\n| CLI | Install |\n|-----|---------|\n| [Claude Code](https://docs.anthropic.com/en/docs/claude-code/overview) | `npm install -g @anthropic-ai/claude-code` |\n| [Gemini CLI](https://github.com/google-gemini/gemini-cli) | `npm install -g @google/gemini-cli` |\n| [Codex CLI](https://github.com/openai/codex) | `npm install -g @openai/codex` |\n| [OpenCode](https://opencode.ai) | `curl -fsSL https://opencode.ai/install | bash` |\n\n> Why two? Because one AI talking to itself is a monologue, not a collaboration.\n\n---\n\n## Manual Installation\n\nPrefer to install per-client yourself? Each command is one line.\n\n### Claude Code\n\n```bash\nclaude mcp add --scope user Multi-CLI -- npx -y @osanoai/multicli@latest\n```\n\n<details>\n<summary>Claude Desktop (JSON config)</summary>\n\nAdd to `~/Library/Application Support/Claude/claude_desktop_config.json` on macOS:\n\n```json\n{\n  \"mcpServers\": {\n    \"Multi-CLI\": {\n      \"command\": \"npx\",\n      \"args\": [\"-y\", \"@osanoai/multicli@latest\"]\n    }\n  }\n}\n```\n\nRestart Claude Desktop completely after saving.\n</details>\n\n---\n\n### Gemini CLI\n\n```bash\ngemini mcp add --scope user Multi-CLI npx -y @osanoai/multicli@latest\n```\n\n<details>\n<summary>Manual config (~/.gemini/settings.json)</summary>\n\n```json\n{\n  \"mcpServers\": {\n    \"Multi-CLI\": {\n      \"command\": \"npx\",\n      \"args\": [\"-y\", \"@osanoai/multicli@latest\"]\n    }\n  }\n}\n```\n</details>\n\n---\n\n### Codex CLI\n\n```bash\ncodex mcp add Multi-CLI -- npx -y @osanoai/multicli@latest\n```\n\n<details>\n<summary>Manual config (~/.codex/config.toml) or pass --mcp-config</summary>\n\n```bash\ncodex --mcp-config mcp.json\n```\n\nWhere `mcp.json` contains:\n\n```json\n{\n  \"mcpServers\": {\n    \"Multi-CLI\": {\n      \"command\": \"npx\",\n      \"args\": [\"-y\", \"@osanoai/multicli@latest\"]\n    }\n  }\n}\n```\n</details>\n\n---\n\n### OpenCode\n\nOpenCode's `mcp add` command is interactive, so add Multi-CLI to `~/.config/opencode/opencode.json` directly:\n\n```json\n{\n  \"mcp\": {\n    \"Multi-CLI\": {\n      \"type\": \"local\",\n      \"command\": [\"npx\", \"-y\", \"@osanoai/multicli@latest\"]\n    }\n  }\n}\n```\n\nIf the file already exists, merge the `\"Multi-CLI\"` entry into the existing `\"mcp\"` object.\n\n---\n\n### Any Other MCP Client\n\nMulti-CLI uses standard stdio transport. If your client supports MCP, point it at:\n\n```\nnpx -y @osanoai/multicli@latest\n```\n\n---\n\n## Available Tools\n\nOnce connected, your AI client gains access to tools for the *other* CLIs (never its own):\n\n| Tool | Description |\n|------|-------------|\n| `List-Gemini-Models` | List available Gemini models and their strengths |\n| `Ask-Gemini` | Ask-Gemini a question or give it a task |\n| `Fetch-Chunk` | Retrieve chunked responses from Gemini |\n| `Gemini-Help` | Get Gemini CLI help info |\n| `List-Codex-Models` | List available Codex models |\n| `Ask-Codex` | Ask-Codex a question or give it a task |\n| `Codex-Help` | Get Codex CLI help info |\n| `List-Claude-Models` | List available Claude models |\n| `Ask-Claude` | Ask-Claude a question or give it a task |\n| `Claude-Help` | Get Claude Code CLI help info |\n| `List-OpenCode-Models` | List available OpenCode models from all configured providers |\n| `Ask-OpenCode` | Ask-OpenCode a question or give it a task |\n| `OpenCode-Help` | Get OpenCode CLI help info |\n\n---\n\n## Usage Examples\n\nOnce installed, just talk naturally to your AI:\n\n```\n\"Ask-Gemini what it thinks about this architecture\"\n\"Have Codex review this function for performance issues\"\n\"Get Claude's opinion on this error message\"\n\"Use OpenCode to get a second opinion from Llama\"\n```\n\nOr get a second opinion on anything:\n\n```\n\"I want three perspectives on how to refactor this module —\n ask Gemini and Codex what they'd do differently\"\n```\n\n---\n\n## How It Works\n\n```\n┌─────────────┐     MCP (stdio)      ┌──────────────┐     CLI calls    ┌─────────────┐\n│  Your AI    │ ◄──────────────────► │  Multi-CLI   │ ───────────────► │ Other AIs   │\n│  Client     │                      │   server     │                  │ (CLI tools) │\n└─────────────┘                      └──────────────┘                  └─────────────┘\n\n1. Your AI client connects to Multi-CLI via MCP\n2. Multi-CLI detects which CLIs are installed on your system\n3. It registers tools for the OTHER clients (hides tools for the calling client)\n4. When a tool is called, Multi-CLI executes the corresponding CLI command\n5. Results flow back through MCP to your AI client\n```\n\n---\n\n## Troubleshooting\n\n**\"No usable AI CLIs detected\"**\nMake sure at least one other CLI is installed and on your PATH:\n```bash\nwhich gemini && which codex && which claude && which opencode\n```\n\n**No tools showing up?**\nIf only your own CLI is installed, Multi-CLI hides it (no self-calls). Install a *different* CLI to enable cross-model collaboration.\n\n**MCP server not responding?**\n1. Check that Node.js >= 20 is installed\n2. Run `npx @osanoai/multicli@latest` directly to see if it starts\n3. Restart your AI client completely\n\n---\n\n## Development\n\n```bash\ngit clone https://github.com/osanoai/multicli.git\ncd multicli\nnpm install\nnpm run build\nnpm run dev\n```\n","readmeFilename":"README.md","_rev":"1-d3fb6a642e3f30c9419ea8761c3701a4"}