使用说明
neuralingual-mcp
CLI for Neuralingual — AI-powered affirmation practice sets.
Create personalized affirmation sets from natural language intent, render them as audio with premium voices, and manage your library from the command line.
Install
npm install -g @innerstacklabs/neuralingual-mcp
Or run directly with npx:
npx @innerstacklabs/neuralingual-mcp login
Quick Start
# Log in with Apple Sign-In (opens browser)
neuralingual login
# On a remote host (ssh, container) — prints a URL + code to approve on any device
neuralingual login --device
# Create a practice set from intent text
neuralingual create "I want to feel confident before presentations"
# Browse your library
neuralingual library
# Get detailed info on a set
neuralingual info <id>
# Configure rendering
neuralingual voices # browse available voices
neuralingual render configure <id> --voice <voice-id> --context general --duration 10
# Render and play
neuralingual render start <id> --wait
neuralingual play <id> --open
Commands
| Command | Description |
|---|---|
login |
Log in via Apple Sign-In (browser; --device for headless hosts) |
logout |
Log out and clear tokens |
whoami |
Show current user info |
library |
List your practice sets |
search <query> |
Search by title or context |
create <text> |
Create a new practice set |
info <id> |
Show detailed set info |
rename <id> |
Rename a practice set |
delete <id> |
Delete a practice set |
render configure |
Configure render settings |
render start |
Start a render job |
render status |
Check render progress |
rerender <id> |
Re-render with current config |
voices |
List available voices |
voices preview <id> |
Preview a voice |
play <id> |
Download/play rendered audio |
download <job-id> |
Download audio as MP3 |
credits |
Show credit balance |
share <id> |
Generate a share link |
unshare <id> |
Revoke a share link |
set export <id> |
Export set as YAML |
set edit <id> |
Edit set in $EDITOR |
set apply <id> |
Apply YAML changes |
set create |
Create set from YAML |
settings |
View/update preferences |
Logging in on a remote host (ssh, containers)
Over ssh or in a container, the browser that opens is on a different machine than the
terminal, so the normal Apple Sign-In flow can't complete. neuralingual login detects this
automatically (any ssh session, or a Linux host with no DISPLAY/WAYLAND_DISPLAY) and
switches to a device-code flow — pass --device to force it, or --browser to force the
loopback flow instead (passing both is an error).
Open https://app.neuralingual.com/auth/device and enter code ABCD-EFGH
Or open this link directly: https://app.neuralingual.com/auth/device?code=ABCD-EFGH
Waiting for approval (the code expires in 10 minutes)...
Open the URL on any device — phone, laptop, whatever's handy — sign in with Apple, and
approve. The CLI polls in the background and writes tokens to
~/.config/neuralingual/auth.json, the same file the browser flow writes, so nothing else
about the CLI changes once you're logged in. The code expires after 10 minutes; if it times
out, the CLI exits non-zero and tells you to run neuralingual login again.
See the User Guide for detailed usage and examples.
MCP Server (AI Assistant Integration)
This package includes an MCP server that lets AI assistants (Claude Code, etc.) manage your Neuralingual library directly.
Setup
Add to your .mcp.json (Claude Code) or MCP client config:
{
"mcpServers": {
"neuralingual": {
"type": "stdio",
"command": "npx",
"args": ["-y", "-p", "@innerstacklabs/neuralingual-mcp", "neuralingual-mcp"]
}
}
}
Prerequisite: Run neuralingual login first to authenticate.
MCP Tools (44 tools)
Library & Discovery
| Tool | Description |
|---|---|
nl_library |
List all practice sets with title, context, render status |
nl_library_list |
List playlists with play count and last-played metadata |
nl_library_view |
View a playlist with affirmations, feedback, and play stats |
nl_info |
Full set details — affirmations, render config, share status |
nl_search |
Search sets by keyword |
nl_guide |
Render the framework (methodology, principles, sources) as markdown |
Creation & Editing
| Tool | Description |
|---|---|
nl_create |
Create a new set from intent text and/or source material (costs 1 credit) |
nl_rename |
Update title and/or emoji |
nl_delete |
Delete a practice set |
nl_intent_update |
Update intent text, title, emoji, or tone |
nl_intent_delete |
Delete a playlist (alias for nl_delete) |
nl_generate_more |
Generate additional affirmations for an existing set (costs 1 credit) |
Affirmation Management
| Tool | Description |
|---|---|
nl_affirmation_add |
Add a custom affirmation to a playlist |
nl_affirmation_delete |
Delete a single affirmation |
nl_affirmations_feedback |
Like or dislike affirmations (batch) |
nl_affirmations_toggle |
Enable or disable affirmations (batch) |
nl_sync_affirmations |
Declarative sync — add, update, remove affirmations |
nl_set_export |
Export set as editable YAML (optionally with framework) |
nl_set_import |
Apply edited YAML back to a set |
Audio & Rendering
| Tool | Description |
|---|---|
nl_voices |
List available voices with accent, gender, tier |
nl_render_configure |
Configure voice, background, pace, duration, context |
nl_render_start |
Start audio rendering |
nl_render_status |
Check render progress |
nl_rerender |
Re-render with current config |
nl_play |
Download rendered audio to local cache |
nl_playback_start |
Log the start of a practice session |
nl_playback_complete |
Record session duration and completion |
Source Extraction
| Tool | Description |
|---|---|
nl_source_extract |
Extract text from a web article URL |
nl_source_youtube |
Extract transcript from a YouTube video |
nl_source_twitter |
Extract text from a Twitter/X post |
nl_source_pdf |
Upload and extract text from a local PDF |
Catalog
| Tool | Description |
|---|---|
nl_catalog_browse |
Browse curated catalog with filtering and sorting |
nl_catalog_view |
View a catalog item with full details |
nl_catalog_copy |
Copy a catalog playlist to your library |
Settings & Preferences
| Tool | Description |
|---|---|
nl_context_settings_list |
List per-context default settings |
nl_context_settings_update |
Update defaults for a session context |
nl_context_settings_reset |
Reset a context to system defaults |
nl_wizard_defaults |
Get resolved wizard defaults (voice, background, binaural, etc.) |
Account
| Tool | Description |
|---|---|
nl_user_profile |
Get user profile, subscription tier, and credits |
nl_user_set_username |
Set or update username |
nl_user_check_username |
Check username availability |
nl_credits |
Check credit balance |
nl_share |
Generate a public share link |
nl_unshare |
Revoke a public share link |
Example (Claude Code)
> Create a neuralingual set about being a great father
Creating "Present & Intentional Dad" with 30 affirmations...
Drew on Fred Rogers, Brené Brown, and John Gottman.
> Render it with Graham's voice, meditation context, 15 minutes
Configured: Meditation · Graham · Acoustic Guitar · 15m
Rendering... done.
> Play it
Downloaded to ~/.config/neuralingual/audio/abc123.mp3
License
MIT
