Command-line client and MCP server for Glossa — analyze text and manage your vocabulary from the terminal or any MCP-aware LLM client
Project description
glossa-cli
A standalone command-line client for Glossa — analyze text and manage your vocabulary from the terminal.
You'll need a Glossa account — sign up at glossa.pro. For setup help and the full integration guide, see glossa.pro/account.
Install (one command)
pipx install glossa-cli # recommended — isolated install
# or:
pip install glossa-cli
Both put two binaries on your PATH:
glossa— the terminal CLI for vocabulary managementglossa-mcp— a stdio MCP server for Claude Desktop, ChatGPT, etc.
Sign in
Run glossa login first. It opens a browser to authorise you; after you
approve, your tokens are cached in ~/.config/glossa/credentials.json and
refreshed automatically — you won't need to sign in again.
glossa login # opens a browser; prints an auth URL to approve
glossa login --force # clear the cached sign-in and authorise again
glossa logout # sign out — remove cached credentials
Switching accounts is glossa logout && glossa login (or just
glossa login --force). The glossa-mcp server reads the same credentials
file, so re-auth on the CLI then restart your MCP client to pick up the
new sign-in.
CI / scripting: skip the browser flow by exporting a static token from Glossa → Settings → Tokens:
export GLOSSA_API_TOKEN=glossa_xxx.
Quick start
glossa login # one-time browser sign-in
glossa add serenity # add a word to your vocabulary
glossa list --status learning # see what you're learning
glossa review # interactive FSRS review
echo "We've done the impossible, and that makes us mighty." | glossa analyze # CEFR breakdown
If your account has multiple learning languages enabled, Glossa will ask you
to pick one. Set a default with glossa lang set <code> (persistent), or
export GLOSSA_LANG=<code> (session). Single-language accounts don't need to
think about this.
MCP server
The same install ships glossa-mcp, a stdio MCP server for Claude Desktop,
ChatGPT, and other MCP-aware LLM clients. Add it to your client's config —
full setup guide is in the Glossa Account → MCP tab. To smoke-test:
glossa-mcp
# Run interactively → prints a brief help message explaining how to use it.
# Run with piped JSON-RPC on stdin → serves the protocol on stdout.
Usage
# Analyze text (arg, --file, or stdin) — language auto-detected
echo "And we choked 'em with those words." | glossa analyze
glossa analyze --file article.txt
glossa analyze "a short sentence" --json
# Add a word or phrase
glossa add serenity --status learning
glossa add "hold my beer" -t "watch how it's done"
# Change a word's state (upserts by lemma; --id targets one sense)
glossa set serenity known
glossa set --id 1234 saved
# List your vocabulary
glossa list --status learning
glossa list --json
# Interactive FSRS review session
glossa review
# Default learning language (only needed if your account has 2+ enabled)
glossa lang # show current default + source
glossa lang set fr # persist to ~/.config/glossa/config.json
glossa lang unset
Project details
Release history Release notifications | RSS feed
Download files
Download the file for your platform. If you're not sure which to choose, learn more about installing packages.
Source Distributions
Built Distribution
Filter files by name, interpreter, ABI, and platform.
If you're not sure about the file name format, learn more about wheel file names.
Copy a direct link to the current filters
File details
Details for the file glossa_cli-0.1.1-3-py3-none-any.whl.
File metadata
- Download URL: glossa_cli-0.1.1-3-py3-none-any.whl
- Upload date:
- Size: 25.6 kB
- Tags: Python 3
- Uploaded using Trusted Publishing? No
- Uploaded via: uv/0.8.19
File hashes
| Algorithm | Hash digest | |
|---|---|---|
| SHA256 |
b41ab9df1dde3a6e77aa95f2e698e489b22bc6989e5aa13905a03425ba90530d
|
|
| MD5 |
5053147c80a3952f180ead42a8819e3c
|
|
| BLAKE2b-256 |
1cac20486f9a4d8943737bda9cc055acbd1f07c614b9661dee05062e057cde4e
|