unb-consultant

Universal NotebookLM-Based Consultant. Create and query AI experts powered by Google Gemini Notebook (formerly NotebookLM).

Installation

pipx install "unb-consultant[browser]"

Or with cookie extraction support:

pipx install "unb-consultant[browser,cookies]"

Quick Start

# Authenticate
unb login

# Create an expert
unb expert create "cvss-v4" \
  --url "https://www.first.org/cvss/v4-0/specification-document" \
  --desc "CVSS v4.0 scoring expert"

# Ask a question
unb ask "cvss-v4" "How is the base score calculated?"

# List experts
unb expert list

MCP Server

For AI agent integration, start the MCP server:

unb mcp

Then configure in your opencode.json:

{
  "mcpServers": {
    "unb-consultant": {
      "command": "unb",
      "args": ["mcp"]
    }
  }
}

Commands

Command	Description
`unb login`	Authenticate with Google
`unb auth check --test`	Verify authentication
`unb auth refresh`	Refresh authentication tokens
`unb expert create NAME`	Create a new expert (`--url`, `--file`, `--auto`)
`unb expert list`	List all experts (`--json` for structured output)
`unb expert delete NAME`	Delete an expert (`--yes` to skip confirmation)
`unb ask NAME QUESTION`	Ask a question (`--json` for citations)
`unb suggest KEYWORD ...`	Suggest expert domains by keywords
`unb catalog NAME`	Generate thematic catalog
`unb skill-gen NAME`	Generate SKILL.md (`--auto` for automatic)
`unb source add NAME`	Add sources to expert
`unb init`	Set up unb-consultant in a project (`--auto`)
`unb setup`	Interactive setup wizard
`unb mcp`	Start MCP server

Development

Prerequisites

Python 3.10+
Google account (for NotebookLM)

Local setup

git clone https://github.com/freddyeleazar/unb-consultant.git
cd unb-consultant
pip install -e ".[browser,dev]"

Testing

pytest

Releasing

See AGENTS.md for the release workflow. The project uses Conventional Commits and Keep a Changelog.

Metadata

Release files for unb-consultant 0.2.0

For a detailed explanation of source distributions (sdists) and built distributions (wheels), please see the package formats documentation.

Source distribution (sdist)

Source distribution for unb-consultant 0.2.0
File	Size	Uploaded
unb_consultant-0.2.0.tar.gz	35.3 kB	Jul 26, 2026	Details

Built distribution (wheel)

Table of built distributions (wheels) for unb-consultant 0.2.0
File	Interpreter	ABI	Platform	Reset
unb_consultant-0.2.0-py3-none-any.whl 35.9 kB Jul 26, 2026	Python 3	none	any	Details

Total release size: 71.2 kB

Release files / unb_consultant-0.2.0.tar.gz

Download URL	unb_consultant-0.2.0.tar.gz
Size	35.3 kB
Tags	Source
SHA-256 checksum How to use checksums	`25fe000fa10cd8ee4276a121072223c9ef10f4aed89ef53739aa01813a986eb8`
BLAKE2b-256 checksum How to use checksums	`77248231216c6e8cfee35d7b2ff650404889819ee0adbce7b5d3727333efacef`
Upload date	Jul 26, 2026
Uploaded using Trusted Publishing? What is trusted publishing?	Yes
Uploaded via	`twine/6.1.0 CPython/3.13.14`

Provenance

Provenance describes where a file came from. On PyPI, provenance is shared via attestations, which provide a verifiable record of the build or publishing details. View details, limitations and caveats.

Identity

Publishing platform: GitHub Actions
Publishing repository: github.com/freddyeleazar/unb-consultant
Publishing commit: github.com/freddyeleazar/unb-consultant/tree/18cb8a15a1388bbce3158305e28817d7e2b18f5e

Workflow

Publishing configuration: github.com/freddyeleazar/unb-consultant/blob/18cb8a15a1388bbce3158305e28817d7e2b18f5e/.github/workflows/publish.yml
Publishing logs: github.com/freddyeleazar/unb-consultant/actions/runs/30187319571/attempts/1

Artifact

Subject: unb_consultant-0.2.0.tar.gz
SHA-256 checksum: 25fe000fa10cd8ee4276a121072223c9ef10f4aed89ef53739aa01813a986eb8 Verifying attestations

Release files / unb_consultant-0.2.0-py3-none-any.whl

Download URL	unb_consultant-0.2.0-py3-none-any.whl
Size	35.9 kB
Tags	Python 3
SHA-256 checksum How to use checksums	`91bb0fcbbca37ec4ca5fe82f1d8ed888d9ced9b6fe56da55e0b0bbf1d298b3cb`
BLAKE2b-256 checksum How to use checksums	`ebb408f60bd5093f7f60f9c18a761a9cc91ebe7d5bb3da91907104f26e7516fe`
Upload date	Jul 26, 2026
Uploaded using Trusted Publishing? What is trusted publishing?	Yes
Uploaded via	`twine/6.1.0 CPython/3.13.14`

Provenance

Identity

Publishing platform: GitHub Actions
Publishing repository: github.com/freddyeleazar/unb-consultant
Publishing commit: github.com/freddyeleazar/unb-consultant/tree/18cb8a15a1388bbce3158305e28817d7e2b18f5e

Workflow

Publishing configuration: github.com/freddyeleazar/unb-consultant/blob/18cb8a15a1388bbce3158305e28817d7e2b18f5e/.github/workflows/publish.yml
Publishing logs: github.com/freddyeleazar/unb-consultant/actions/runs/30187319571/attempts/1

Artifact

Subject: unb_consultant-0.2.0-py3-none-any.whl
SHA-256 checksum: 91bb0fcbbca37ec4ca5fe82f1d8ed888d9ced9b6fe56da55e0b0bbf1d298b3cb Verifying attestations