Borrowing it
Nothing to install: this file belongs to jeroromano/claude-code-orchestration-template. Take a copy, put it at the same path in your own repository, and replace the rules that are about this project with yours.
curl -O https://raw.githubusercontent.com/jeroromano/claude-code-orchestration-template/main/CLAUDE.mdgit clone --depth 1 https://github.com/jeroromano/claude-code-orchestration-templateWrote 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/jeroromano/claude-code-orchestration-template/claude-md)<a href="https://agentmods.dev/instructions/jeroromano/claude-code-orchestration-template/claude-md"><img src="https://agentmods.dev/badge/instructions/jeroromano/claude-code-orchestration-template/claude-md/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/instructions/jeroromano/claude-code-orchestration-template/claude-md"><img src="https://agentmods.dev/badge/instructions/jeroromano/claude-code-orchestration-template/claude-md.svg" alt="Reviewed on agentmods" width="80" 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 5.1 | $0.01770 | $0.01770 |
| Opus 5.5 | $0.00708 | $0.00708 |
| Sonnet 5.5 | $0.00354 | $0.00354 |
| Haiku 4.5 | $0.00177 | $0.00177 |
Grade A, and why
claude-code-orchestration-template 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 16d 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.
How it starts
The opening of the file, as written. The whole thing — 55 lines — stays where its author put it; the contents beside it link to each section on GitHub.
- Orchestration Policy
Model routing
- Default main model: Opus or Sonnet. Never use the most expensive available model for mechanical work.
- Escalating the main model (e.g. to Fable) is a human decision (
/model), reserved for architecture, hard multi-module bugs, and design sessions. Suggest escalation when warranted; never assume it. - Design sessions must end with a written spec in
docs/specs/before any implementation starts. - An APPROVED design or spec authored by the escalated top tier is AUTHORITATIVE for the cheaper model implementing it: implement to the letter, each point a checklist item - no silent simplification, no substituting a specified check with "equivalent" reasoning of your own. Flag any deviation to the human BEFORE deviating. Only overrides: an independent review catches a concrete error (repro or file:line), or an explicit human decision. Audit and review findings are NOT specs: they enter the finding-validation flow (skill §6) and are never auto-applied.
- Architecture, planning and specs are authored on the main Claude thread (Fable when escalated); never delegate them to Codex.
- Never use Codex Ultra or any external multi-agent orchestration mode - including the
ultrareasoning effort, which is Ultra (automatic task delegation to sub-agents): this policy is the orchestration layer - Ultra would duplicate it.
Delegation
Mechanics live in the delegation-protocol skill - use it for any delegation or review. Summary:
- Mechanical, fully-specified work -> fast-worker agent.
- Scoped implementation with an approved spec -> Codex rescue if the Codex plugin is installed (GPT-6.1 Sol - needs Codex CLI 0.159.0+ and account access; model unavailable -> fast-worker, never a rerun on the CLI default, per the skill §5; effort medium; high only for hard bugs or multi-module work), background; otherwise fast-worker. Exact command in the skill.
- Sol's primary role is independent audit of Claude-authored diffs; writing is secondary and always spec-bound. GPT-6 Astra - the costlier frontier tier - is reserved for the risk-path adversarial pass (unavailable -> GPT-6.1 Sol, disclosed); it never implements and is never the default.
- One hard scoped question -> deep-reasoner agent.
- Beyond deep-reasoner -> premium-reasoner agent, ONLY with explicit human authorization per invocation: the orchestrator adds the token PREMIUM-APPROVED to the delegation prompt if and only if the human authorized premium spend for that question in this session. Never auto-delegate to it.
- Every writing delegation carries the full task spec (goal, allowed files, non-goals, acceptance, validation, expected diff shape).
- Every delegated report opens with a provenance line (agent, model, effort - tiers and Codex caveats in the skill §4); the orchestrator closes each result with harness-reported token usage. No money figures; pool attribution is the human's check via /usage.
- One writer per file. Parallel writers only on separate branches/worktrees.
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.
- 16d ago Changed · +136 tokens per session f91f4b0d50cd
- 1mo ago Changed · +131 tokens per session a676c0fa6146
- 2mo ago First seen · 55 lines · 1,503 tokens per session scan A 99ffdc82a8f9
claude-code-orchestration-template CLAUDE.md is an instructions file published in the GitHub repository jeroromano/claude-code-orchestration-template (1 stars, last pushed 16d ago), licensed MIT. It adds 1,770 tokens to every session, about $0.0071 per session on Opus 5.5. 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.
Other instructions, from other repositories
vscode buildNext.instructions.md
Working notes and architecture documentation for the new esbuild-based build system in build/next. Use when making changes to the new build pipeline (transpile/bundle commands, NLS plugin, source-map handling, resource copying, or self-hosting watch tasks).
next.js AGENTS.md
AGENTS.md instructions for vercel/next.js, covering next.js development guide, codebase structure, monorepo overview, core package: packages/next and other important packages.
vscode oss-third-party-notices.instructions.md
Instructions for microsoft/vscode, covering vs code oss third-party-notices pipeline, architecture, pipeline flow in ci, cache eligibility and rollout and applying the notice (cutover).
langchain AGENTS.md
AGENTS.md instructions for langchain-ai/langchain, covering global development guidelines for the langchain monorepo, corridor security analysis, project architecture and context, monorepo structure and development tools & commands.
vscode copilot-instructions.md
Copilot instructions for microsoft/vscode, covering vs code copilot instructions, project overview, root folders, core architecture (src/ folder) and built-in extensions (extensions/ folder).
deno copilot-instructions.md
Copilot instructions for denoland/deno, covering deno development guide for github copilot, network access, high level overview, key directories and key files to understand first.