| FMRS | 46 / 100 · D | 79 / 100 · B |
| Reliability | 8 / 20 | 12 / 20 |
|---|
| Security and permissions | 8 / 20 | 18 / 20 |
|---|
| Maintenance | 9 / 20 | 16 / 20 |
|---|
| Documentation | 12 / 20 | 18 / 20 |
|---|
| Setup experience | 9 / 20 | 15 / 20 |
| Best for | - Ableton Live 12 users who want an AI assistant for music production.
- Producers needing intelligent device recommendations, sample analysis, and composition assistance.
- Musicians who prefer controlling their DAW with natural language or artist/genre shorthand.
- Advanced users who want the system to learn their personal plugin library.
- Studios wanting AI-assisted mixing, sound design, and composition with deep integration.
| - Engineering teams scaling AI coding agents while controlling spend
- Platform teams that need auditable, rollback-aware agent execution records
- Organizations using multiple agents (Claude, Codex, Gemini) under one governed flow
|
| Not for | - Beginners unfamiliar with Ableton Live 12 (requires operating the DAW).
- Users needing fully autonomous AI composition (LivePilot returns plans, not autonomous execution).
- Users who only need simple MIDI control without complex tooling.
- Those expecting a free commercial plugin (it uses BSL-1.1 license).
- Users on other DAWs (Logic, FL Studio) — it only supports Ableton Live 12.
| - Developers who just want a simple code-completion helper without run governance
- Teams unwilling to accept local JSONL run records with HMAC signing
- Environments without a supported adapter (Claude/Codex/Gemini CLIs, direct provider, or verifier-only)
|
| Required permissions | - Communicates with Ableton Live 12 over TCP 9878 (requires Remote Script installation).
- Communicates with the Max for Live analyzer over UDP 9880 and OSC 9881 (optional, for spectral tools).
- Reads local Splice database files and communicates with the Splice desktop app over gRPC (optional).
- Scans the filesystem for plugins and samples (requires user directory access).
- Writes state files to `~/.livepilot/`, including atlas overlays, memory, and logs.
- Makes changes to the Live session (tempo, tracks, devices) — all through the undo stack.
| - Read/write access to the configured repo root (writable scope restrictable via --allow-path/--deny-path)
- Ability to invoke the chosen agent CLI (Claude/Codex/Gemini, etc.) and its underlying model account
- Local disk writes for run records, receipts, and share outputs (share/run-receipt., etc.)
- Execution of user-configured verifier commands (e.g. npm test)
|
| Risks and side effects | - Modifies the Live session in real time, which may inadvertently change your project; backup recommended.
- Splice interactions can consume credits on downloads; there's a safety floor of 5 credits, but still a risk.
- The M4L analyzer listens on LAN (OSC 9881); unauthenticated commands are dropped since v1.28.0, but use trusted networks.
- Filesystem scans can be time-consuming and expose plugin paths/names (privacy).
- With many tools, mis-calling may lead to complex state changes; experimental in production sessions.
- Actively developed; APIs may change frequently. Pin a version for stable work.
| - Verifier commands are executed; poorly chosen verifier commands carry security risk despite preflight policy checks
- Cost and token figures may be estimates (labeled with provenance) and should not be treated as settled accounting
- Receipt integrity must be verified before use as trustworthy evidence; missing rollback evidence is flagged as EVIDENCE_BOUNDARY
- Policy defaults come from martin.config.yaml and can be overridden by CLI flags; misconfiguration can loosen safety boundaries
|
| Supported clients | Claude Desktop, Claude Code, Cursor, VS Code | Claude Code, Codex, Gemini CLI, Cursor, VS Code, GitHub Copilot, Continue |
| Tools | 10 | 0 |