CLI client for podwise.ai — turn any podcast episode into AI-powered insights, designed for use in AI agents and skills workflows.

Podwise transforms hours of podcasts into transcripts, summaries, outlines, Q&A, and mind maps. This CLI is purpose-built as a tool for AI agents — letting LLMs, skills runtimes, and automation pipelines fetch structured podcast insights without a browser or human in the loop.

• 🤖 Looking for ready-to-use agent skills? Jump to Agent Skills →
• 🔌 Looking for mcp server? Jump to MCP Server →
• 🧩 Prefer a hosted MCP server, or a Claude Code / Codex plugin? See hardhackerlabs/podwise-plugin →

brew install hardhackerlabs/podwise-tap/podwise

Run the following command to install the latest version of podwise:

curl -sL https://raw.githubusercontent.com/hardhackerlabs/podwise-cli/main/install.sh | sh

1. Download the latest binary for your OS and architecture from GitHub Releases.
2. Unpack the archive (e.g., tar -xzf podwise_linux_amd64.tar.gz).
3. Move the podwise binary to a directory in your PATH, for example:

   mv podwise /usr/local/bin/

4. Make sure it's executable: chmod +x /usr/local/bin/podwise.

If you have Go installed, you can build and install the binary directly from the source:

git clone https://github.com/hardhackerlabs/podwise-cli.git
cd podwise-cli
go build -o podwise .
# Move the binary to a directory in your PATH, e.g.,
mv podwise /usr/local/bin/

Authorize the CLI in your browser and let it save the API key automatically:

# Open the browser authorization flow
podwise auth

# Verify connection
podwise config show

If you already have an API key, you can still set it manually:

podwise config set api_key your-sk-xxxx

The configuration is stored at ~/.config/podwise/config.toml.

podwise popular

# Ask a question answered from podcast transcripts
podwise ask "the future of AI Agents"

# Search episodes by title keywords
podwise search episode "machine learning" --limit 20

# Search podcasts by name
podwise search podcast "Lex Fridman"

# Podwise episode URL (Recommended)
podwise process https://podwise.ai/dashboard/episodes/7360326

# 小宇宙 episode URL 
podwise process https://www.xiaoyuzhoufm.com/episode/abc123

# Youtube video URL
podwise process https://www.youtube.com/watch?v=d0-Gn_Bxf8s

# local file
podwise process ./episode.mp3

# Get summary
podwise get summary https://podwise.ai/dashboard/episodes/7360326

# Get transcript
podwise get transcript <episode-url>

For more details on all available commands and flags, run:

podwise --help

Prerequisites: Before installing skills, make sure you have completed the Installation and Configuration steps above — the podwise CLI must be installed and your api_key must be set.

The Podwise skill routes your intent to the right workflow automatically:

When you ask the agent something related to podcasts, it:

1. Loads Podwise CLI reference docs
2. Matches your intent to the right workflow (if applicable)
3. Loads and executes the workflow steps, or runs the CLI command directly

npx skills add hardhackerlabs/podwise-cli

You can also build your own skills on top of the podwise CLI to create custom workflows that fit your needs.

Hosted alternative — no local process. The same Podwise MCP server is also available as a remote endpoint at https://mcp.podwise.ai/mcp (Streamable HTTP, OAuth 2.1), packaged together with the Podwise agent skill in hardhackerlabs/podwise-plugin. That plugin is listed in Anthropic's plugin directory and on the official MCP Registry as io.github.hardhackerlabs/podwise (server.json). It exposes 38 tools, including search_episodes, get_episode_transcript, get_episode_summary, ask_podwise, translate_episode and export_episode_markdown.

Podwise exposes all its capabilities as an MCP (Model Context Protocol) server, allowing AI assistants (Gemini CLI, Claude Desktop, Cursor, etc.) to search episodes, process media, and retrieve AI-generated content directly as tools.

Start the MCP server (communicates over stdin/stdout):

podwise mcp

The server exposes the following tools:

1. Make sure podwise is installed and configured (see Installation and Configuration).

2. Add the following to your settings file:

{
  "mcpServers": {
    "podwise": {
      "command": "podwise",
      "args": ["mcp"]
    }
  }
}