Hasab MCP

Model Context Protocol (MCP) server for Hasab AI. Lets Cursor, Claude Desktop, and other MCP hosts call Hasab language APIs from natural language—translation, summarization, transcription, text-to-speech, and chat.

Install from PyPI; calls https://api.hasab.ai with your HASAB_KEY. Source repository is private to Hasab.

Docs: logic.md (architecture & releases) · code.md (implementation)

Requirements

• Python 3.11+
• uv recommended (uvx)
• Hasab API key (HASAB_KEY_… from the portal)

Install

pip install hasab-mcp
# or one-off / always latest cache:
uvx hasab-mcp

PyPI: https://pypi.org/project/hasab-mcp/

Cursor

Add to ~/.cursor/mcp.json (or project .cursor/mcp.json):

{
  "mcpServers": {
    "hasab": {
      "command": "uvx",
      "args": ["hasab-mcp"],
      "env": {
        "HASAB_API_KEY": "HASAB_KEY_…"
      }
    }
  }
}

Restart Cursor. You should see six tools: hasab_translate, hasab_summarize, hasab_transcribe, hasab_tts, hasab_list_tts_speakers, hasab_chat.

Example prompts

• Translate "Good morning" to Amharic using Hasab.
• Using Hasab TTS, speak "ሰላም" in Amharic with speaker hanna. Give me the audio file path.

Claude Desktop

Same mcpServers block in:

• macOS: ~/Library/Application Support/Claude/claude_desktop_config.json
• Windows: %APPDATA%\Claude\claude_desktop_config.json

Tools (v1)

Translation language codes: amh, eng, orm, tig (and aliases en, am, ti, Oromo, Tigrigna).

Environment

Do not commit API keys. Each user uses their own key; usage bills to their Hasab account.

License

Proprietary — Hasab AI. All rights reserved. Not open source. Internal and customer use via published PyPI package only.