eBird CLI

A command-line interface for exploring eBird bird observation data.

Prerequisites

• Python 3.11+
• eBird API Key (obtainable with free eBird account)

Installation

1. Clone the repository
2. Set up a virtual environment
3. Install the ebird-cli command from the checkout:

   pip install .
   

4. Configure your eBird API key

Configuration

Command-Line Arguments

The CLI supports the following arguments:

Environment Variables

You can configure the CLI using the following environment variables to avoid manual parameter entry:

• EBIRDAPIKEY: Your eBird API key
• EBIRDDEFAULTREGION: Default region for default search and hotspots filtering (CA-QC-MR)
• EBIRDLOCALE: Preferred language locale (fr)
• EBIRDLAT: Latitude for location-based searches (47.87)
• EBIRDLONG: Longitude for location-based searches (-72.17)
• EBIRDYEARLIST: Path to your year observations list (~/ebird_data/year_list.csv)
• EBIRDLIFELIST: Path to your lifetime observations list (~/ebird_data/life_list.csv)
• EBIRDAGGREGATIONS: Path to your aggregations JSON file (~/ebird_data/aggregations.json)

Usage

Launching the CLI

ebird-cli --api-key <ebird-api-key> --region <region-code> [optional arguments]

ebird-cli --version prints the installed version.

Available Commands

When launched, the CLI will display a menu of available commands:

• recent <scope, -region> [-back, -species]: Fetch recent bird observations
• notable <scope, -region> [-back]: Fetch notable bird observations

Autocompletion

Suggestions appear as you type: command names, scopes, flags, region names for the chosen scope, species names and -back values. Press Tab to accept the first suggestion, or press it again to cycle through them.

Search Scopes

Bird observations can be searched in multiple scopes.

1. Nearby: Based on provided latitude and longitude

   • Requires --lat and --long program arguments
   • Shows observations near the specified geographic coordinates
   • recent nearby

2. Hotspot: eBird-defined hotspots

   • recent hotspot -region Dunes de Tadoussac

3. Regional: eBird Subnational Level 2

   • recent regional -region Montréal

4. Subnational: eBird Subnational Level 1

   • recent subnational -region Québec

The -region flag supports context-sensitive autocompletion based on the selected scope. If no -region flag is provided, default region will be used.

Search length

Use the optional -back parameter to change the number of days back to fetch observations.

• Values: from 1 to 30

• Default: 7

  notable subnational -region Québec -back 30
  

Aggregations

Aggregations allow you to define custom named groups of hotspots that appear alongside eBird hotspots when using the hotspot scope. This is useful for combining multiple nearby locations under a single name.

Provide the path to your aggregations file via --aggregations or the EBIRDAGGREGATIONS environment variable:

ebird-cli --api-key <key> --region <region> --aggregations ~/ebird_data/aggregations.json

The aggregations file is a JSON array where each element maps a custom name to a list of eBird location IDs:

[
  {
    "Technoparc": ["L9980102", "L4886900", "L11363355"]
  }
]

Once configured, aggregation names are available in the -region autocomplete when using the hotspot scope:

recent hotspot -region Technoparc

Species filter

Use the optional -species flag with the recent command to filter observations to a single species.

recent nearby -species Canard colvert
recent regional -region Montréal -species Harfang des neiges

The -species flag supports context-sensitive autocompletion using the cached taxonomy. Start typing a common name and press Tab to see matching species.

Note: Autocompletion and lookup use the locale specified by --locale / EBIRDLOCALE. Species names must be typed in the configured locale.

List highlighting

When using --year-list and --life-list:

• Year list targets are highlighted in green
• Life list targets are highlighted in red

Obtaining list files

The eBird API does not expose life list or year list data, so these CSV files must be downloaded from the eBird website. The ebird-cli-download-lists command, installed alongside ebird-cli, downloads both into output_dir as life_list.csv and year_list.csv:

ebird-cli-download-lists session_id output_dir

This requires a valid eBird session cookie (EBIRD_SESSIONID), which can be obtained from your browser after logging in to ebird.org. The command must be re-run whenever your lists change.

Note: The life list only changes when you see a species for the first time ever, making it infrequent and easy to manage. The year list changes with every new species seen in the current year and resets annually, requiring more frequent re-runs of the command.

Data cache

To avoid redundant API calls to eBird, the CLI caches reference data locally on first run. The cache is stored in the platform-appropriate user cache directory (e.g. %LOCALAPPDATA%\ebird_cli\Cache\ebird_cli on Windows, ~/.cache/ebird_cli on Linux/macOS).

The following data is cached:

• Location data (per region): subnational level 1 regions, subnational level 2 regions, and every hotspot in the subnational region — stored as JSON files under location/<national>/ and location/<national>/<subnational>/
• Taxonomy (per locale and national region): the full eBird taxonomy filtered to species observed in the configured country — stored as CSV files under taxonomy/<national>/

Hotspots are fetched again on launch once their cached file is more than 30 days old, so new hotspots appear; if eBird can't be reached, the older file is used. All other cached data is written once and reused on every subsequent launch. To force a refresh (e.g. after a new eBird taxonomy release or a region change), delete the relevant files or the entire cache directory.

Log file

Each launch appends debug messages to app.log in the platform-appropriate user log directory, created on first launch:

• Windows: %LOCALAPPDATA%\ebird_cli\ebird_cli\Logs\app.log
• Linux: ~/.cache/ebird_cli/log/app.log (under $XDG_CACHE_HOME instead of ~/.cache when it is set)
• macOS: ~/Library/Logs/ebird_cli/app.log

The file is never rotated or trimmed; delete it at any time.

Development setup

Virtual environment

1. Activate the virtual environment:

   • On Windows: venv\Scripts\activate
   • On macOS/Linux: source venv/bin/activate

2. Install the project in editable mode with the dev tools (requirements-dev.txt adds the test, lint and build tools to the pinned runtime dependencies):

   pip install -e . -r requirements-dev.txt
   

   The tests read the version from the installed package, so they need this editable install.

Tests

The test suite runs offline: it never calls the eBird API, needs no API key and never touches your user cache or log directory.

python -m pytest

Tests live in tests/, mirroring the source layout:

• tests/cli/: the recent and notable commands and autocompletion, driven through their public entry points
• tests/services/, tests/domain/: cache, location and taxonomy services and domain value parsing
• tests/utils/: where the app log is written
• tests/test_prompt_dispatch.py, tests/test_program_arguments.py: prompt input handling and program arguments
• tests/conftest.py: shared fixtures, including make_cli, which wires real services and commands the way main does
• tests/fakes.py: the in-memory fake eBird client, the only test double
• tests/fixtures/: small hand-written cache, list and aggregation files for the CA-QC-MR region

Tests named ..._currently_... pin behaviour that looks accidental, so changing it is a deliberate test change.

Demo GIFs

The GIFs in this README are recorded with VHS from the tapes in docs/demo/. They call the live eBird API, so they need your API key and environment variables set. config.tape holds the shared settings and launches the CLI out of frame.

On Windows, install VHS and its terminal dependency, then re-record a GIF from the repository root:

winget install charmbracelet.vhs tsl0922.ttyd

vhs docs/demo/overview.tape

VHS also needs ffmpeg on the PATH.

Cutting a release

Releases are published to PyPI by the Release workflow (.github/workflows/release.yml) through trusted publishing; no API token is involved.

1. Bump version in pyproject.toml and merge it to main.
2. Optionally, do a dry run: run the Release workflow manually from the Actions tab. It uploads that version to TestPyPI, which accepts each version only once. Check the upload installs:

   pip install --index-url https://test.pypi.org/simple/ --extra-index-url https://pypi.org/simple/ ebird-cli
   

3. Tag the release commit vX.Y.Z, matching the version, and push the tag.
4. Publish a GitHub release from that tag. The workflow uploads it to PyPI, and fails without uploading if the tag doesn't match the version in pyproject.toml.

Credits

Bird observation data and the eBird platform are provided by the Cornell Lab of Ornithology. This project would not be possible without their freely available eBird API.

This project relies on the following open-source packages:

• ebird-api — Python wrapper for the eBird API
• prompt_toolkit — Interactive command-line interface with autocompletion
• rich — Terminal formatting and highlighted output
• pandas — CSV observation list parsing
• colorama — Cross-platform terminal color support
• requests — HTTP requests to the eBird API

License

MIT License