CORE: Instructions file for Claude Code

CLAUDE.md

CORE CLAUDE.md is an instructions file for Claude Code from DariuszNewecki/CORE. It costs 1,943 tokens per session, scanned A, original, MIT.

A set of project instructions for Claude Code, an AI programming assistant, explaining CORE’s structure, rules, and working process.

In plain words
What is it for?
Use it when working on CORE, a software factory that supervises AI-generated code under human-defined rules and verification.
Why use it?
It tells the assistant which files define requirements, architecture, and implementation boundaries before it changes code.

Instructions file for Claude Code

Written for Claude Code: the file is CLAUDE.md. Also seen: mentions CLAUDE.md.

This is DariuszNewecki/CORE's own configuration. It tells Claude Code how to work on CORE itself, so it is not a mod to install elsewhere. Copy it as a starting point and replace the rules that are about this project. Everything CORE configures →

Reuse

Borrowing it

Nothing to install: this file belongs to DariuszNewecki/CORE. Take a copy, put it at the same path in your own repository, and replace the rules that are about this project with yours.

Copy the file
curl -O https://raw.githubusercontent.com/DariuszNewecki/CORE/main/CLAUDE.md
Clone the repo
git clone --depth 1 https://github.com/DariuszNewecki/CORE

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 CORE CLAUDE.md

README.md
[![agentmods](https://agentmods.dev/badge/instructions/dariusznewecki/core/claude-md/github.svg)](https://agentmods.dev/instructions/dariusznewecki/core/claude-md)
Your own site
<a href="https://agentmods.dev/instructions/dariusznewecki/core/claude-md"><img src="https://agentmods.dev/badge/instructions/dariusznewecki/core/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.

agentmods 80×15 button for CORE CLAUDE.md

Your own site · 80×15
<a href="https://agentmods.dev/instructions/dariusznewecki/core/claude-md"><img src="https://agentmods.dev/badge/instructions/dariusznewecki/core/claude-md.svg" alt="Reviewed on agentmods" width="80" height="20"></a>
Per session 1,943 This file is loaded in full into every session.
When invoked 1,943 The same file — it is already loaded in full.
Security scan A 0 findings. A grade says what 26 rules found in the file — not that it is safe.
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 $0.01943 $0.01943
Opus 5.5 $0.00777 $0.00777
Sonnet 5.5 $0.00389 $0.00389
Haiku 4.5 $0.00194 $0.00194

Measured 5d ago against content hash 2709830de344, method: parsed. Prices are Anthropic first-party input rates as of 2026-10-07, from the pricing page.

Security

Grade A, and why

CORE 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 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.

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 · 136 lines

How it starts

The opening of the file, as written. The whole thing — 136 lines — stays where its author put it; the contents beside it link to each section on GitHub.

CLAUDE.md — CORE

v3 done; CORE 3.1 in progress (since 2026-10-29)

The freeze ended 2026-10-29: every v3 gate issue is closed (milestone CORE 3.0: 48 closed, 0 open). Work is filed in milestone CORE 3.1. No new features before the v3 release. New infrastructure (timers, crons, credentials, services) needs the governor's yes first. The PR check freeze-scope still refuses a PR whose description names no v3 or CORE 3.1 issue.

CORE is a Python runtime that audits a codebase against rules in .intent/, records violations in PostgreSQL, proposes fixes, and logs the chain from finding to commit.

Layout

src/api/     FastAPI routes; DB access only via api/dependencies.py
src/body/    atomic actions, services, workers (execution)
src/mind/    rule engines; read .intent/, no I/O or execution
src/will/    agents and autonomous orchestration
src/shared/  common substrate; must not import mind/body/will
src/cli/     Typer commands; Rich rendering only here
.intent/     rules and config the runtime reads (YAML/JSON)
.specs/      design history (ADRs, papers); reference, not instructions

Facts (update when they change)

  • PyPI: core-runtime 3.1.0 (2026-11-03; consumer: proof run 10).
  • Latest proof run: 9 (#1053, 2026-10-30, on 3.0.0 from PyPI); gate met, no issues filed. Next run: 10.
  • Production-ready (docs/release.md): stays Beta until an independent person, not an AI agent, installs CORE from the README and uses it.
  • Credentials: a closed list (governor 2026-10-23). No new token, no permission change, never asked of the governor. A job uses one of these and checks it at start, failing loudly if it does not work.
    • core-claude-bot gh login (core-claude; scopes repo, workflow, read:org): pushes branches, opens PRs into main and turns on their auto-merge, creates promote/<sha> branches, opens or updates the promotion PR, and closes a superseded one and deletes its branch. After review finding F2: also opens the PR for CORE's generated tests. The 06:00 job hands this login's token to promote propose as CORE_GITHUB_TOKEN. Every 15 minutes infra/scripts/core_alert_check.py opens and closes the [alert] issues that mention the governor (review F4); no new credential.
    • CORE's own token (CORE daemon): read as CORE_GITHUB_TOKEN from the environment, else the unit credential github-token (delivery_branch.py). Pushes CORE's branches; cannot open PRs (#1252). Same variable name as above, a different credential.
    • /etc/core/pipeline-github-token (core-runner): an issues-only token on the governor's own account (#1198); it opens the issue for a failed promotion, and every other pipeline read is unauthenticated. GitHub sends no email for one's own issue, so an alert meant to reach the governor is opened as core-claude-bot (#1201 was opened as the governor).
    • /etc/core/pipeline-deploy-key (core-runner): the write deploy key; moves live. Ruleset live-writers lets any write deploy key do so; this is the only one (governor counted, 2026-10-23), and it stays the only one.
    • Deliberately not on the list: the governor's own GitHub login, the built-in GITHUB_TOKEN of Actions jobs, and PyPI (Trusted Publishing, no stored token).

Read the full file on GitHub · 136 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. 5d ago Changed · +7 tokens per session 2709830de344
  2. 7d ago Changed · +4 lines · +96 tokens per session a59014019cfe
  3. 8d ago Changed · +2 lines · +27 tokens per session ec6d25b0cfab
  4. 9d ago Changed b3669a60346c
  5. 10d ago Changed · -1 lines · -37 tokens per session a5f815e58e85
  6. 16d ago Changed · +41 lines · +755 tokens per session 8fbe4b688ca2
  7. 20d ago Changed · +6 tokens per session 91909371cb4b
  8. 21d ago Changed · +24 lines · +343 tokens per session 373f712ecdf0
  9. 26d ago Changed · -478 lines · -8,121 tokens per session scan B → A f4ae8362ec83
  10. 27d ago First seen · 544 lines · 8,867 tokens per session scan B fd058a96f9d6

Subscribe to this mod's changes

CORE CLAUDE.md is an instructions file published in the GitHub repository DariuszNewecki/CORE (39 stars, last pushed yesterday), licensed MIT. It adds 1,943 tokens to every session, about $0.0078 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 2026-08-30.

Related

CORE CLAUDE.md compared with instructions from other repositories

CORE CLAUDE.md next to 6 instructions from other repositories
ModTypeStarsLast pushLicenceScanTokensWorks with
CORE CLAUDE.mdDariuszNewecki/COREInstructions file39yesterdayMITA1,943Claude Code
vscode buildNext.instructions.mdmicrosoft/vscodeInstructions file195ktodayMITA9,719GitHub Copilot
next.js AGENTS.mdvercel/next.jsInstructions file143kyesterdayMITA7,461Codex, OpenCode
vscode oss-third-party-notices.instructions.mdmicrosoft/vscodeInstructions file195ktodayMITA5,649GitHub Copilot
langchain AGENTS.mdlangchain-ai/langchainInstructions file149ktodayMITA4,680Codex, OpenCode
vscode copilot-instructions.mdmicrosoft/vscodeInstructions file195ktodayMITA3,093GitHub Copilot
vscode ai-customization.instructions.mdmicrosoft/vscodeInstructions file195ktodayMITA1,907GitHub Copilot