nexus-agents CLAUDE.md

nexus-agents CLAUDE.md is an instructions file for Claude Code from nexus-substrate/nexus-agents. It costs 13,092 tokens per session, scanned A, original, MIT.

A project instruction set for AI coding agents, defining their rules, procedures, and official file locations. It also describes review and audit practices for tracking agent work.

In plain words
What is it for?
Following project governance, locating canonical documentation, reviewing agent changes, and recording work for later checks.
Why use it?
It gives different coding agents one shared source of instructions, reducing missed rules and differences between tools.

Instructions file for Claude Code

Written for Claude Code: the file is CLAUDE.md. Also seen: mentions CLAUDE.md; mentions subagents; names the AskUserQuestion tool.

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/nexus-substrate/nexus-agents/claude-md
Clone the repo
git clone --depth 1 https://github.com/nexus-substrate/nexus-agents

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 nexus-agents CLAUDE.md

README.md
[![agentmods](https://agentmods.dev/badge/instructions/nexus-substrate/nexus-agents/claude-md.svg)](https://agentmods.dev/instructions/nexus-substrate/nexus-agents/claude-md)
Your own site
<a href="https://agentmods.dev/instructions/nexus-substrate/nexus-agents/claude-md"><img src="https://agentmods.dev/badge/instructions/nexus-substrate/nexus-agents/claude-md.svg" alt="Measured on agentmods" height="20"></a>
Per session 13,092 This file is loaded in full into every session.
When invoked 13,092 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 5.1 $0.13092 $0.13092
Opus 5 $0.06546 $0.06546
Sonnet 5 $0.02618 $0.02618
Haiku 4.5 $0.01309 $0.01309

Measured today against content hash b91c03e10196, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-06, from the pricing page.

Security

Grade A, and why

nexus-agents 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 today.

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 · 564 lines

How it starts

The opening of the file, as written. The whole thing — 564 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Nexus Agents - Claude Code Instructions

Project: Governance substrate for AI coding agents — adversarial review, drift-detected rules, tamper-evident append-only audit chain (threat model: tamper-evident, not tamper-proof), closed-loop telemetry. The agents (Claude/Codex/Gemini/OpenCode) do the engineering; nexus-agents enforces the rules they have to follow, reviews their work adversarially, and audits everything they touch. Repository: github.com/nexus-substrate/nexus-agents Owner: @williamzujkowski


Where the rest lives. Everything harness-agnostic — the CLI quick reference, the NEXUS_* environment table, Operating Rules, the Discovered-Issues protocol, Workflows, Governance quality, File References and the MCP tool list — now lives in AGENTS.md and is injected into the generated block above (#5151). It used to sit in this file only, so Codex, Gemini CLI and OpenCode never saw it. Content outside the generated block is ungated by construction, which is how the TypeScript pin drifted.

Mission

Nexus-agents is a self-improving autonomous engine for an organization's technical work, built to extend toward its broader functions over time. It runs the engineering lifecycle end to end — form/refine intent → research → plan (agile PM: epics → child issues) → develop (multi-CLI) → adversarial-consensus review → QA + security gates → measure outcomes → self-tune. The routing/strategy self-improvement loop is wired and default-on (outcome telemetry → strategy distillation → bounded routing adjustment, enforce-by-default since v2.96). Coding was the deliberate bootstrap: build an engine good enough to extend itself to the remaining functions and keep improving how it works.

Read the full file on GitHub · 564 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. today Changed · +1 lines · +91 tokens per session b91c03e10196
  2. 2d ago Changed faf66708e629
  3. 5d ago Changed · +11 lines · +299 tokens per session 7b503fbe3b3c
  4. 6d ago First seen · 552 lines · 12,702 tokens per session scan A 9cfb6983ca95

Subscribe to this mod's changes

nexus-agents CLAUDE.md is an instructions file published in the GitHub repository nexus-substrate/nexus-agents (18 stars, last pushed yesterday), licensed MIT. It adds 13,092 tokens to every session, about $0.0655 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.