local-copilot

local-copilot is a skill for Claude Code from Pantani/ableton-mind. It costs 110 tokens per session (638 once invoked), scanned A, original, MIT.

A set of instructions for working on the local AI assistant feature in ableton-mind. It covers Ollama and OpenAI-compatible services, chat, command-line questions, tool permissions and the browser interface.

In plain words
What is it for?
Adding or debugging local chat and ask features, AI models and endpoints, tool safety levels, loopback-only browser access, documentation and tests.
Why use it?
It keeps the assistant read-only by default and makes failures clear when Ableton Live is unavailable. It also sets boundaries around write actions, browser access and sensitive tools.

Skill for Claude Code

Written for Claude Code: installed under .claude/.

Install

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.

agentmods
npx agentmods add skills/pantani/ableton-mind/local-copilot
Any agent
npx skills add Pantani/ableton-mind --skill local-copilot
Clone the repo
git clone --depth 1 https://github.com/Pantani/ableton-mind

Made for: Claude Code.

Wrote 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.

agentmods badge for local-copilot

README.md
[![agentmods](https://agentmods.dev/badge/skills/pantani/ableton-mind/local-copilot.svg)](https://agentmods.dev/skills/pantani/ableton-mind/local-copilot)
Your own site
<a href="https://agentmods.dev/skills/pantani/ableton-mind/local-copilot"><img src="https://agentmods.dev/badge/skills/pantani/ableton-mind/local-copilot.svg" alt="Measured on agentmods" height="20"></a>
Per session 110 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.
When invoked 638 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. Scan, not verified.
Origin original No closer match found in the catalogue.
Token cost

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.

ModelPer sessionOnce invoked
Fable 5.1 $0.00110 $0.00638
Opus 5 $0.00055 $0.00319
Sonnet 5 $0.00022 $0.00128
Haiku 4.5 $0.00011 $0.00064

Measured 5d ago against content hash a2b8f95f3144, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-05, from the pricing page.

Security

Grade A, and why

local-copilot scanned grade A with 0 findings 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 5d 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.

Nothing flagged

None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.

.claude/skills/local-copilot/SKILL.md · 56 lines

How it starts

The opening of the file, as written. The whole thing — 56 lines — stays where its author put it; the contents beside it link to each section on GitHub.

local-copilot

Use this skill when adding, debugging or extending the ableton-mind local LLM feature.

Context Check

  1. Read src/llm/config.ts, src/llm/tools.ts, src/llm/agent.ts, src/llm/server.ts and src/cli/chat.ts.
  2. Read tests/llm-local-copilot.test.ts and the latest relevant QA report.
  3. Check _workspace/tdmcp-compatible-features.md for deferred tdmcp-inspired items.
  4. Confirm whether the request is UI chat, headless ask, model/backend, safety tier, docs, or QA.

Safety Rules

  • Keep the default tier read-only (safe). Use explicit --write for simple mutations and --creative for recipes/browser load.
  • Never expose destructive or broad routing/mixer tools to a local model without a safety review.
  • Preserve the bridge-offline behavior: the chat UI can open without Live; live tools must return clear bridge errors.
  • Browser server stays loopback-only and must reject non-loopback Host/Origin.
  • Do not copy TouchDesigner-specific prompts, paths or tools from tdmcp.

Implementation Rules

  • Prefer explicit JSON Schemas in src/llm/tools.ts over new schema-conversion dependencies.
  • The local copilot should use existing ToolDefinition handlers, not duplicate Ableton business logic.
  • Add or update tests whenever changing config defaults, tool tier membership, agent loop behavior, CLI parsing or server endpoints.
  • Public docs belong in English root docs and matching Portuguese pages under docs/pt/.

Verification

Run focused checks first:

npm run typecheck
npx vitest run tests/llm-local-copilot.test.ts

For broader changes, add:

npm run lint
npm run test
npm run build

Live/Ollama validation is optional unless the task changes runtime connectivity. If skipped, report it explicitly.

Test Scenarios

  • Normal: ableton-mind ask --read-only "What is in this set?" with Ollama reachable and bridge connected.
  • Offline bridge: ableton-mind chat --no-ollama --no-open opens the local UI even when Live is closed; live tool calls return a bridge-offline error.
  • Safety: default config resolves to safe; --write switches to standard; --creative exposes recipes/browser load.

Read the full file on GitHub · 56 lines

Changes

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.

  1. 5d ago First seen · 56 lines · 0 tokens per session scan A a2b8f95f3144

Subscribe to this mod's changes

local-copilot is a skill published in the GitHub repository Pantani/ableton-mind (5 stars, last pushed 2mo ago), licensed MIT. It adds 110 tokens to every session and 638 once invoked, about $0.0006 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.

Related

Other skills, from other repositories

ableton-live

Connect and operate the Loophole Bridge (Ableton Live MCP server). Use when the user wants to check Live/bridge prerequisites, wire an MCP client to Loophole, or run a Live editing recipe (humanize, build arrangement, batch rename, chord from prompt). Triggers: "is my Ableton bridge working", "set up loophole"…

OthmanAdi/loophole · 90 tokens

groove-builder

Use when the user wants drum patterns by genre - kick, snare, hi-hat, percussion. Examples - "give me a trap beat", "house drum pattern", "DnB drums at 174", "lo-fi drums with swing", "boom-bap pattern".

glincker/ableton-skills · 60 tokens

midi-cleanup

Use when the user asks to humanize a MIDI part, fix voice leading, quantize notes, clean up timing, fix stuck/duplicate notes, or generally polish a MIDI clip they recorded or wrote. Examples - "humanize this hat", "voice these chords better", "the timing is too stiff", "fix the voice leading on the strings".

glincker/ableton-skills · 76 tokens

mixer-doctor

Use when the user describes a mix problem ("muddy", "harsh", "no headroom", "vocals get lost", "kick and bass fighting") or asks for a mix audit. Diagnoses the issue from session state and proposes specific corrective moves with EQ, compression, sends, and routing.

glincker/ableton-skills · 67 tokens

producer-mode

Use when the user asks to set up tracks, pick instruments, scaffold an arrangement, build a project template, or describe a track they want to make in Ableton Live. Examples - "make me a 4-bar lo-fi loop", "set up a film score template", "I want to start a hip-hop beat in C minor".

glincker/ableton-skills · 72 tokens

reference-match

Use when the user references another track they want to emulate the vibe/sound of. Examples - "make this sound like Ólafur Arnalds", "I want a Tame Impala vibe", "match the mix of Kendrick's Money Trees". Translates artist/track references into actionable production decisions.

glincker/ableton-skills · 65 tokens