Project
Oremi Kànsà
MCP server for podcast discovery and playback.
33 commits 1 contributor Updated Nov 2026
Overview
Oremi Kànsà is an MCP server for discovering podcasts through the Podcast Index. It provides AI assistants and autonomous agents with a standardized interface to search, browse, and retrieve podcast information using the Model Context Protocol (MCP).
Designed to run in Docker, Oremi Kànsà exposes tools for searching podcasts, exploring episodes, browsing categories, and retrieving metadata from Podcast Index. It enables MCP-compatible clients to access podcast content without interacting directly with the Podcast Index API.
The name Oremi Kànsà comes from the Yoruba language. Oremi means "my friend", reflecting the project's goal of building a helpful and trustworthy AI ecosystem. Kànsà is derived from the Yoruba word associated with broadcasting or broadcast media, representing the service's responsibility for helping AI agents discover podcast content through a unified MCP interface.
Core Capabilities
- Search and discover podcasts by topic, language, and category
- Retrieve detailed podcast and episode metadata
- Browse categories, trending shows, and related podcasts
- Return structured, validated response models for MCP clients
- Use either the public Podcast Index API or the offline SQLite catalog
- Accelerate searches with persistent FTS5 indexes and a bounded memory cache
- Persist and periodically refresh the catalog in
/oremi/datain Docker - Run with multiple Docker Swarm replicas against the same cluster-shared volume
- Configure database updates, HTTP timeouts, HTTP binding, API prefix, and logging through environment variables
- Serve MCP over Streamable HTTP
- Run as a lightweight Docker container with simple deployment.
- Integrate seamlessly with the Oremi ecosystem and other MCP-compatible clients.
Project Links
Technology
- AI
- ai-agents
- audio
- llm
- mcp
- media
- oremi
- podcast
- Python
- streaming
Repository information
- Primary language
- Python
- Created
- July 16, 2026
- Last activity
- November 16, 2026
Activity
Daily commits on the default branch, across the whole project history.