HermeticOrmus

Opus Magnum — AI skill for Claude Code

AI community

The Great Work for Claude Code — an adaptive method for long-horizon, high-trust AI builds.

How to install Opus Magnum

This entry records only its repository, not the path inside it, so there is no exact command to give. Open HermeticOrmus/opus-magnum and copy the folder into ~/.claude/skills/, or the file into ~/.claude/agents/.

What Opus Magnum does

The Great Work for Claude Code — an adaptive method for long-horizon, high-trust AI builds. Triage, elements, modes; Quintessence journal, Ouroboros loop, cross-model Assay, guardrails.

Alternatives in AI

Bmad Method — Breakthrough Method for Agile Ai Driven Development 42712 382 2 42.9k ★
LongHorizon Harness — The long-horizon computer-use harness 1.3k ★
Late CLI — High-performance AI agent for long-horizon tasks and local models 442 ★

README

Opus Magnum Claude Code

The Great Work for Claude Code — an adaptive method for long-horizon, high-trust AI builds

A method, a skill, and an architecture for running very long Claude Code builds — attended or unattended — without losing context, quality, or control.

Why this exists

Long AI coding sessions fail in predictable ways: context windows degrade, agents drift from intent, "green tests" hide empty work, and nobody is watching when it goes wrong. Opus Magnum is the answer distilled from months of real long-run practice: the "long session" is never one heroic context window. It is a durable journal plus a loop of fresh, primed contexts, bounded by guardrails and gated by a second model's review.

The method is adaptive. Most tasks do not need the whole machine — they need two or three of its elements, applied deliberately.