ORCH CLAUDE.md

Project instructions for ORCH, an AI agent runtime that lets software agents work together on engineering tasks. They describe its architecture, commands, testing, and release process.

In plain words
What is it for?
Use them when building, testing, type-checking, understanding the project structure, or preparing a release of ORCH.
Why use it?
They give a coding agent the project’s rules and accurate commands, reducing incorrect changes and inconsistent development work.

Instructions file

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 instructions/oxgeneral/orch/claude-md
Clone the repo
git clone --depth 1 https://github.com/oxgeneral/ORCH
Per session 1,725 This file is loaded in full into every session.
When invoked 1,725 The same file — it is already loaded in full.
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 $0.01725 $0.01725
Opus 5 $0.00863 $0.00863
Sonnet 5 $0.00345 $0.00345
Haiku 4.5 $0.00172 $0.00172

Measured 2d ago against content hash 468f7be22338, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

ORCH CLAUDE.md 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 2d 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.md · 149 lines

How it starts

The opening of the file, as written. The whole thing — 149 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.

Positioning

ORCH is an AI agent runtime — not just a CLI tool. The core engine (domain + application + infrastructure) has zero dependencies on CLI/TUI layers. The npm package @oxgeneral/orch exports the full engine API via src/index.ts, enabling programmatic use from any Node.js application.

User-facing messaging: "Turn your AI tools into a coordinated engineering team" Technical category: "AI agent runtime" Internal architecture term: "ORCH Engine"

Commands

npm run build          # Build ESM + DTS bundles (tsup)
npm run dev            # Run via tsx (no build needed)
npm test               # Run all tests (vitest)
npm test -- test/unit/application/orchestrator-resilience.test.ts  # Single file
npm test -- --grep "state machine"                                  # By pattern
npm run typecheck      # tsc --noEmit (strict mode)
npm run coverage       # Tests with coverage report

Release: ./scripts/release.sh patch|minor|major && git push && git push --tags

Architecture

Layered DDD with dependency injection — no frameworks, no decorators.

Domain (models, state machine, errors)
  ↓
Application (services, orchestrator, event bus)
  ↓
Infrastructure (adapters, file storage, process manager, templates, workspaces)
  ↓
CLI (Commander.js) / TUI (Ink/React)

Container (DI)

src/container.ts defines two tiers:

  • LightContainer: stores + services only — used by read-only commands (task, agent, logs, config, context, msg, goal, team, status)
  • Container (extends Light): + orchestrator + adapters + ProcessManager + LiquidJS — used by run, tui, doctor

The CLI entry point (src/bin/cli.ts) determines which tier to build based on the subcommand, then lazy-imports only the needed command module.

State Machine

src/domain/transitions.ts — tasks flow through: todo → in_progress → review → done, with retrying and failed branches. All transitions validated by canTransition(). Terminal statuses: done, failed, cancelled.

Read the full file on GitHub · 149 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. 2d ago First seen · 149 lines · 1,725 tokens per session scan A 468f7be22338

Subscribe to this mod's changes

ORCH CLAUDE.md is an instructions file published in the GitHub repository oxgeneral/ORCH (155 stars, last pushed 1mo ago), licensed MIT. It adds 1,725 tokens to every session, about $0.0086 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 2026-08-30.