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 instructions/federico-pepe/ableton-push-hack/claude-mdgit clone --depth 1 https://github.com/federico-pepe/ableton-push-hackWrote 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/instructions/federico-pepe/ableton-push-hack/claude-md)<a href="https://agentmods.dev/instructions/federico-pepe/ableton-push-hack/claude-md"><img src="https://agentmods.dev/badge/instructions/federico-pepe/ableton-push-hack/claude-md.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 | $0.08031 | $0.08031 |
| Opus 5 | $0.04015 | $0.04015 |
| Sonnet 5 | $0.01606 | $0.01606 |
| Haiku 4.5 | $0.00803 | $0.00803 |
Grade A, and why
ableton-push-hack CLAUDE.md scanned grade A 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 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.
Makes network callslowCapability
Not a fault in itself. Listed so you know the mod talks to something, and to what.
| `push-catalog.sh` (embedded into the binary via `make embed`) | All the logic: `q()`/`rq()` (python3-only JSON readers — no jq dependency, mirrors the framework installer's own avoidance of it; `q`'s `catalog` op also How it starts
The opening of the file, as written. The whole thing — 256 lines — stays where its author put it; the contents beside it link to each section on GitHub.
CLAUDE.md
This file provides guidance to Claude Code (claude.ai/code) when working with code in this repository.
Doc sync rule: Keep this file, all
docs/files, andREADME.mdin sync with every code change. If a change affects behaviour, APIs, architecture, or known issues — update the relevant docs in the same commit. Save all implementation plans to/plans/with filename formatYYYY-MM-DD-title-of-the-plan.md. Update CHANGELOG.md's## [Unreleased]section in the same commit as any change worth noting to a future reader — new behavior, a fix, a changed API. Skip internal refactors and trivial edits.
Project
push-hack — extensible hack framework for Ableton Push 3 (Intel Linux, runs full Ableton Live). Deploys via SSH. Never modifies system partition. Only three hacks are core — built from this repo, installed directly via install.sh: Push Manager (web file browser + display control, port 7701), Push Display (LD_PRELOAD display hook), Push Hack Catalog (on-device installer for community hacks, port 7702). Everything else is optional and installed via Push Hack Catalog rather than built from this repo — see catalog/catalog.json — including Automation (LFO/CC curve sequencer, port 7703), Browser Bridge (Live MIDI Remote Script to load .adv/.adg presets — one-time manual activation required), and Keyboard Visualizer (on-screen piano keyboard sourced from Live's post-transform notes, port 7705).
core/ is a shared Go module (see "Core shared library" below) that push-manager, automation and keyboard-visualizer all depend on via require+replace — extracted per discovery/push-core-refactor.md to kill the ALSA/HTTP/SSE triplication that had silently diverged across the three hacks.
Core constraint: Push is a live performance tool. Hacks must not crash it, hog CPU, or consume significant memory.
⛔ Hard safety rules — never violate, no exceptions:
- Never modify
/boot/— bricks Push. - Never modify
/opt/— read-only system partition. Contains Push3 app, firmware, assets. - Never modify kernel parameters — no
sysctl -w, no/proc/sys/writes affecting stability. - Never write to
/etc/except:/etc/udev/rules.d/99-push-hack-*.rules,/etc/init.d/push-hack-*, and the LD_PRELOAD line in/etc/init.d/push3(all managed by install/uninstall scripts).
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.
- 5d ago First seen · 256 lines · 8,031 tokens per session scan A 27e179d2ab4b
ableton-push-hack CLAUDE.md is an instructions file published in the GitHub repository federico-pepe/ableton-push-hack (24 stars, last pushed yesterday), licensed MIT. It adds 8,031 tokens to every session, about $0.0402 per session on Opus 5. A static security scan graded it A with 1 finding (makes network calls). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.
Other instructions, from other repositories
cml-mcp AGENTS.md
AGENTS.md instructions for xorrkaz/cml-mcp, covering agents.md — cml mcp server, project overview, compatibility goal, repository layout and tool modules (src/cmlmcp/tools/).
brilliant_sdk AGENTS.md
Instructions for brilliantlabsAR/brilliant_sdk, covering brilliant sdk — agent guide, how an app works (the pattern behind everything), minimal reading paths, verify without hardware and testing.
NeoMind CLAUDE.md
Claude Code instructions for camthink-ai/NeoMind, covering neomind — edge ai platform for iot, development commands, ecosystem repositories, extension package contract (.nep) and device type template contract (json).
cad-cae-copilot copilot-instructions.md
Copilot instructions for armpro24-blip/cad-cae-copilot, covering github copilot — aieng workspace and essentials.
phone-mcp CLAUDE.md
Claude Code instructions for premex-ab/phone-mcp, covering claude.md, project overview, build commands, architecture and module layout.
zmk-config AGENTS.md
AGENTS.md instructions for urob/zmk-config, covering customization guide, ground rules, how the multi-board layout works, adding a new board and where to change what.