ableton-mcp

Overview Schema Source Related Servers Discussions Admin

plan_audio_midi_sketch

Read-onlyIdempotent

Propose approximate monophonic MIDI notes from source audio in Ableton clip beats, using loop or marker region. Does not render pitch, groove, or create MIDI.

Instructions

Propose approximate monophonic source-audio MIDI notes in native warped clip beats. Uses current loop/markers or an explicit region, 0.256–30 seconds, one selected channel and assigned uniform velocity. Does not render clip pitch, groove or effects, change audio, or create MIDI.

Input Schema

Name	Required	Description
`clipId`	Yes	list_clips Session ID or supported list_arrangement_clips ID.
`trackId`	Yes	ID from list_tracks.
`endBeats`	No	Explicit clip-beat end; at most 256 quarter-note beats.
`velocity`	No	Uniform assigned velocity, default 100; not measured dynamics.
`startBeats`	No	Optional explicit clip-beat region start; requires endBeats.
`channelIndex`	No	Selected source channel, zero-based; defaults to 0.

Output Schema

Name	Required	Description	Default
No arguments

Schema Changelog

Changes observed during successful MCP inspections.

Changed3 schema fields changedv0.3.1Oct 23, 2026

changedInput schema / properties / clipId / description

Previous value: -"Session ID from list_clips; Arrangement ID from list_arrangement_clips where supported."New value: +"list_clips Session ID or supported list_arrangement_clips ID."

removedOutput schema / additionalProperties
```
Removed value: -true
```

addedOutput schema / anyOf

Added value: +[
+  {
+    "properties": {
+      "audioCorrectionApplied": {
+        "const": false
+      },
+      "clipId": {
+        "type": "string"
+      },
+      "eligible": {
+        "type": "boolean"
+      },
+      "lengthBeats": {
+        "type": "number"
+      },
+      "notes": {
+        "items": {
+          "properties": {
+            "duration": {
+              "type": "number"
+            },
+            "pitch": {
+              "maximum": 127,
+              "minimum": 0,
+              "type": "integer"
+            },
+            "start": {
+              "type": "number"
+            },
+            "velocity": {
+              "maximum": 127,
+              "minimum": 0,
+              "type": "integer"
+            }
+          },
+          "required": [
+            "pitch",
+            "start",
+            "duration",
+            "velocity"
+          ],
+          "type": "object"
+        },
+        "type": "array"
+      },
+      "reason": {
+        "type": "string"
+      },
+      "sourceFile": {
+        "type": "object"
+      },
+      "sourceWindow": {
+        "type": "object"
+      },
+      "stateVersion": {
+        "minimum": 1,
+        "type": "integer"
+      },
+      "trackId": {
+        "type": "string"
+      }
+    },
+    "required": [
+      "stateVersion",
+      "trackId",
+      "clipId",
+      "eligible",
+      "reason",
+      "lengthBeats",
+      "notes",
+      "sourceFile",
+      "sourceWindow",
+      "audioCorrectionApplied"
+    ],
+    "type": "object"
+  },
+  {
+    "properties": {
+      "error": {
+        "properties": {
+          "message": {
+            "type": "string"
+          },
+          "outcome": {
+            "enum": [
+              "unknown",
+              "not_executed"
+            ]
+          }
+        },
+        "required": [
+          "message",
+          "outcome"
+        ],
+        "type": "object"
+      }
+    },
+    "required": [
+      "error"
+    ],
+    "type": "object"
+  }
+]

Addedv0.3.0Oct 19, 2026

TDQS

A3.8/5.0

Behavior4/5

Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?

Annotations already declare readOnly/idempotent/non-destructive, so the safety profile is covered. The description adds real value beyond them by enumerating exclusions: no pitch, groove or effects modeling, no audio mutation, no MIDI creation — which tells the agent the output is purely a plan to be consumed by another tool.

Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.

Conciseness4/5

Is the description appropriately sized, front-loaded, and free of redundancy?

Two packed sentences, zero filler, with the core action and the exclusion list front-loaded. The phrasing is telegraphic ('one selected channel and assigned uniform velocity'), which is efficient but slightly dense on first read.

Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.

Completeness4/5

Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?

With a rich annotation set, a 100%-documented schema, and an output schema, the description only needs to add scope, constraints and non-goals — which it does. The main remaining gap is that it doesn't say what the agent should do with the result (e.g., that a separate creation step is required).

Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.

Parameters4/5

Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?

Schema coverage is 100%, so the baseline is 3, but the description adds a constraint absent from the schema: the 0.256–30 second region limit (schema only states clip-beat bounds). It also clarifies that channel selection is singular and velocity is an assigned uniform value rather than measured dynamics, reinforcing the schema text.

Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.

Purpose4/5

Does the description clearly state what the tool does and how it differs from similar tools?

Uses a specific verb ('Propose') with a precise resource ('approximate monophonic source-audio MIDI notes') and a stated coordinate system ('native warped clip beats'). This separates it from sibling analysis tools (analyze_audio_clip) and tuning tools (apply_monophonic_audio_tuning), though it never names an alternative outright.

Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.

Usage Guidelines3/5

Does the description explain when to use this tool, when not to, or what alternatives exist?

It states the inputs it consumes ('current loop/markers or an explicit region') and hard limits (0.256–30 seconds, one channel, uniform velocity), so the calling conditions are inferable. But there is no explicit when-to-use statement or comparison against siblings like propose_audio_transient_warp or apply_monophonic_audio_tuning.

Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.

Deploy Server