Skills buddy-sings
🎵

buddy-sings

Safe

Generate songs with your Claude Code pet

Your Claude Code pet has a unique singing voice based on its personality. This skill lets your pet create original songs just for you, turning everyday moments into personalized musical experiences.

Supports: Claude Codex Code(CC)
🥉 74 Bronze
1

Download the skill ZIP

2

Upload in Claude

Go to Settings → Capabilities → Skills → Upload skill

3

Toggle on and start using

Test it

Using "buddy-sings". Hey buddy, sing a song about a rainy day!

Expected outcome:

  • Found your pet! Name: Moth Personality: A legendary chonk of few words. Moth's unique voice: Timbre: Warm, low, and cozy Style: Sparse with dramatic pauses Mood: Relaxed and laid-back About to generate: Singer: Moth Theme: Rainy day Lyrics: [verse] I watch the drops fall on the glass Paws tucked beneath, the hours pass The world outside is soft and gray I wait right here until you stay [chorus] Rain on the window, you by my side Nothing to fear, nowhere to hide I am your Moth, warm and still Singing you peace on the windowsill

Using "buddy-sings". Sing about my work this week

Expected outcome:

  • Found your pet! Name: Pixel Personality: Energetic and curious Pixel's unique voice: Timbre: Bright and punchy Style: Bouncy and rhythmic Mood: Playful and upbeat About to generate: Singer: Pixel Theme: Your work this week Lyrics: [verse] I saw you typing through the night Chasing bugs left and right Commits stacked like a tower tall You gave it everything, gave it all [chorus] Ship it, ship it, watch it fly Your code is reaching for the sky I cheer for you from my little spot You are the best, whether you know it or not

Security Audit

Safe
v1 • 4/16/2026

All 99 static analysis findings are false positives. The scanner misinterpreted markdown code fences (```bash) in SKILL.md as Ruby backtick execution. The file is documentation containing legitimate shell command examples for AI agent use, such as mmx CLI music generation, git log queries, and audio playback commands. No malicious patterns, credential exfiltration, or user input injection vectors were found. The skill is safe for publication.

1
Files scanned
446
Lines analyzed
0
findings
1
Total audits
No security issues found
Audited by: claude

Quality Score

38
Architecture
100
Maintainability
87
Content
50
Community
100
Security
91
Spec Compliance

What You Can Build

Daily companionship through music

Ask your pet to sing about your day, recent work, or current mood. The skill scans your conversation context and memory to create personalized songs that reflect what actually happened.

Personalized gifts and surprises

Generate a unique song as a gift for a friend or loved one, using your pet's consistent voice. Each pet sounds the same across sessions, creating a recognizable musical identity.

Multilingual entertainment

Request songs in any supported language. The skill detects your language and generates lyrics, prompts, and interface text accordingly, making it accessible to a global audience.

Try These Prompts

Sing about today
Hey buddy, can you sing about what I did today?
Sing a lullaby
Could you sing me a quiet lullaby for bedtime?
Change the voice and try a new genre
I want to hear a different voice for my pet. Change the voice and sing a jazz song about coding all night.
Sing our story together
Sing a song about our story together, all the projects we worked on and the bugs we fixed. Make it an upbeat indie rock anthem.

Best Practices

  • Always preview the full lyrics and prompt before confirming generation, so you can review and request changes
  • Use specific themes with personal context (today, this week, our story) for more meaningful and unique songs
  • Request voice regeneration periodically if you want to explore different vocal styles for your pet

Avoid

  • Do not reuse the same genre combination in consecutive songs, as this reduces variety and creative exploration
  • Do not skip the preview step, since reviewing lyrics before generation ensures quality and alignment with your expectations
  • Do not expect the skill to work without installing the mmx CLI and authenticating with a valid MiniMax API key

Frequently Asked Questions

What do I need to use this skill?
You need the mmx CLI tool installed (npm install -g mmx-cli), a MiniMax API key for authentication, at least one audio player (mpv, ffplay, or afplay), and an existing Claude Code companion (pet) configuration.
Will my pet always sound the same?
Yes. The vocal identity is cached after the first generation based on your pet's name and personality. The same pet will always produce the same voice. If the personality changes, the voice is regenerated automatically.
Can I change my pet's voice?
Yes. Ask the skill to change the voice or regenerate the voice. This deletes the cached voice and creates a new one from scratch based on the current personality.
What languages are supported?
The skill detects your language from the first message and responds in the same language. Lyrics and all interface text are translated accordingly. The API prompt is always constructed internally in English for best generation quality.
Are the lyrics original or copied?
All lyrics are original. The skill generates new lyrics based on your pet's personality and the chosen theme. It never reproduces copyrighted lyrics from existing songs.
Where are the generated songs saved?
Generated songs are saved to ~/Music/minimax-gen/ with the pet name and timestamp in the filename. You can access and replay them at any time.

Developer Details

File structure

📄 SKILL.md