Getting it into your agent
One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.
npx agentmods add commands/glincker/ableton-skills/ableton-debuggit clone --depth 1 https://github.com/glincker/ableton-skillsWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/commands/glincker/ableton-skills/ableton-debug)<a href="https://agentmods.dev/commands/glincker/ableton-skills/ableton-debug"><img src="https://agentmods.dev/badge/commands/glincker/ableton-skills/ableton-debug.svg" alt="Measured on agentmods" height="20"></a>What it costs to keep this loaded
Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5 | $0.00039 | $0.00738 |
| Opus 5 | $0.00019 | $0.00369 |
| Sonnet 5 | $0.00008 | $0.00148 |
| Haiku 4.5 | $0.00004 | $0.00074 |
Grade C, and why
ableton-debug scanned grade C with 1 finding against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 3d ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Recursive force deletehighDestructive command
rm -rf with a variable or a broad path is one typo away from removing the wrong tree.
- Don't run `rm -rf` on anything. How it starts
The opening of the file, as written. The whole thing — 68 lines — stays where its author put it; the contents beside it link to each section on GitHub.
/ableton-debug
Walk through a structured diagnostic when something's broken between Claude/Cursor and Ableton.
Workflow
1. Check MCP server reachability
- Try
get_session_info(lightweight read-only call) - If it times out → MCP server isn't running or socket is dead
- If it returns an error → MCP server is up but Ableton-side script is broken
2. Common failure modes
| Symptom | Likely cause | Fix |
|---|---|---|
| "Could not connect to Ableton" | Remote Script not loaded in Live | Verify AbletonMCP is selected as Control Surface in Live's Preferences > Link, Tempo & MIDI |
| Tools time out after 10s | Socket stuck or Live frozen | Restart Ableton Live; restart MCP server process |
| AbletonMCP not in Control Surface dropdown | Script in wrong directory | Verify __init__.py is in ~/Library/Preferences/Ableton/Live X/User Remote Scripts/AbletonMCP/ (Mac) or equivalent on Windows |
| "Script has expired" | Live's script cache is corrupted | Delete Live X folder in Preferences, restart Live, reinstall script |
| WSL2 connection refused | Wrong host (localhost ≠ Windows host from WSL) | Use host.docker.internal or Windows IP from inside WSL |
| Live 12 doesn't see script | Live 12 changed Remote Scripts directory | Use ~/Documents/Ableton/User Library/Remote Scripts/ instead |
| Smithery install flagged as Trojan | False positive on bundled binary | Install MCP server manually, skip Smithery |
| Tools work but params don't change | Live is in a different view (Arrangement vs Session) | Check current_view; some operations only work in one view |
3. Walk the user through verification
In order, ask the user to confirm:
- ✅ Ableton Live is open and a project is loaded?
- ✅ Preferences → Link, Tempo & MIDI → Control Surface shows "AbletonMCP"?
- ✅ The MCP server process is running (Python process listening on the configured port)?
- ✅ Claude Desktop / Cursor / Codex shows the MCP server as "connected" (not "error")?
If any ❌, fix that step before continuing.
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- 3d ago First seen · 68 lines · 39 tokens per session scan C f4dec3f2e68c
ableton-debug is a command published in the GitHub repository glincker/ableton-skills (22 stars, last pushed 10d ago), licensed MIT. It adds 39 tokens to every session and 738 once invoked, about $0.0002 per session on Opus 5. A static security scan graded it C with 1 finding (recursive force delete). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.
Other commands, from other repositories
beat
Guided beat creation — create a beat from scratch with genre, tempo, and instrumentation choices.
mix
Mixing assistant — analyze and balance track levels, panning, and sends.
sounddesign
Sound design workflow — load instruments and effects, shape parameters for a target sound.
arrange
Guided arrangement — build song structure with sections, transitions, and energy flow.
evaluate
Evaluate recent changes — run the full before/after evaluation loop and recommend keep or undo.
perform
Performance mode — enter a safety-constrained live performance context with energy tracking and safe moves.