AgentX: Instructions file for Claude Code

CLAUDE.md

AgentX CLAUDE.md is an instructions file for Claude Code from jnPiyush/AgentX. It costs 4,243 tokens per session, scanned C, original, Apache-2.0.

A Claude Code instruction file for the AgentX repository. It defines the required quality loop, documents to read before coding, and workflow checks that must pass before work is considered complete.

In plain words
What is it for?
Use it to guide Claude Code through the repository's required preparation, review, iteration, and completion steps.
Why use it?
It prevents agents from editing without first following the repository's review and validation process.

Instructions file for Claude Code

Written for Claude Code: the file is CLAUDE.md. Also seen: reads .claude/ paths; mentions CLAUDE.md; mentions subagents.

This is jnPiyush/AgentX's own configuration. It tells Claude Code how to work on AgentX 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 AgentX configures →

Reuse

Borrowing it

Nothing to install: this file belongs to jnPiyush/AgentX. 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/jnPiyush/AgentX/master/CLAUDE.md
Clone the repo
git clone --depth 1 https://github.com/jnPiyush/AgentX

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 AgentX CLAUDE.md

README.md
[![agentmods](https://agentmods.dev/badge/instructions/jnpiyush/agentx/claude-md.svg)](https://agentmods.dev/instructions/jnpiyush/agentx/claude-md)
Your own site
<a href="https://agentmods.dev/instructions/jnpiyush/agentx/claude-md"><img src="https://agentmods.dev/badge/instructions/jnpiyush/agentx/claude-md.svg" alt="Measured on agentmods" height="20"></a>
Per session 4,243 This file is loaded in full into every session.
When invoked 4,243 The same file — it is already loaded in full.
Security scan C 1 finding. 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 $0.04243 $0.04243
Opus 5 $0.02122 $0.02122
Sonnet 5 $0.00849 $0.00849
Haiku 4.5 $0.00424 $0.00424

Measured 3d ago against content hash fe8d1fa19fc1, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-06, from the pricing page.

Security

Grade C, and why

AgentX CLAUDE.md scanned grade C with 1 finding 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 3d 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.

Recursive force deletehighDestructive command

rm -rf with a variable or a broad path is one typo away from removing the wrong tree.

**Blocked commands**: `rm -rf /`, `git reset --hard`, `drop database`
CLAUDE.md · 271 lines

How it starts

The opening of the file, as written. The whole thing — 271 lines — stays where its author put it; the contents beside it link to each section on GitHub.

CLAUDE.md - Entry Point for Claude Code

This file is the equivalent of .github/copilot-instructions.md for Claude Code. It loads once at session start. Keep it small -- point to detailed docs, don't duplicate them.


Quality Loop Hard Rule (NON-SKIPPABLE)

HARD RULE: Before editing, creating, or deleting any file for a code or docs change, run .agentx/agentx.ps1 loop start -p "<task>" as your ABSOLUTE FIRST tool call. Reading files and running loop status are allowed; mutating the workspace before loop start succeeds is a contract violation. The loop is NOT done until .agentx/agentx.ps1 loop complete -s "<summary>" succeeds, and the subagent review pass must be recorded as a structured verdict on the FINAL iteration (loop iterate ... --verdict approved --reviewer <id> --high 0 --medium 0). loop complete fails on a non-approved verdict, on non-zero HIGH/MEDIUM, or when iterations were recorded after the approval. The pre-commit hook blocks commits when these conditions are unmet.

Honesty rule: If asked whether the loop ran, run .agentx/agentx.ps1 loop status and report the actual state. Do not claim completion unless loop complete succeeded in the current session.

Risk-based iterations: The CLI enforces minimums of standard 1, auto-fix 2, complex/AgentX 3, and high-risk 5. A structured independent reviewer verdict is still required on the final iteration. Report each iteration with loop iterate -s "...", then summarize before completing.

Cross-Cutting Agent Protocol: The shared rules (quality loop, subagent review, per-iteration reporting, Karpathy, Model Council, Scrub, Brainstorm, Plan, Research) are defined ONCE in .github/AGENT-PROTOCOL.md. Agent files keep only the front-loaded Pre-edit gate + Honesty rule stubs and point there.

Frontmatter-only enforcement is insufficient -- this rule lives here in body prose because models routinely skip rules placed only in deeper docs.

Read the full file on GitHub · 271 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. 3d ago First seen · 271 lines · 4,243 tokens per session scan C fe8d1fa19fc1

Subscribe to this mod's changes

AgentX CLAUDE.md is an instructions file published in the GitHub repository jnPiyush/AgentX (15 stars, last pushed yesterday), licensed Apache-2.0. It adds 4,243 tokens to every session, about $0.0212 per session on Opus 5. A static security scan graded it C with 1 finding (recursive force delete). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-09-03.

Related

Other instructions, from other repositories

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.

vercel/next.js · 7,296 tokens

codex AGENTS.md

AGENTS.md instructions for openai/codex, covering rust/codex-rs, the codex-core crate, code review rules, crate api surface and model visible context.

openai/codex · 5,182 tokens

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).

microsoft/vscode · 6,785 tokens

vscode oss-third-party-notices.instructions.md

Instructions for microsoft/vscode, covering vs code oss third-party-notices pipeline, architecture, pipeline flow in ci, applying the notice (cutover) and fallback chain (never fail the build).

microsoft/vscode · 5,001 tokens

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.

langchain-ai/langchain · 4,469 tokens

spec-kit AGENTS.md

AGENTS.md instructions for github/spec-kit, covering agents.md, about spec kit and specify, quickstart — add a new integration in 5 steps, integration architecture and integrationmanifest — file tracking.

github/spec-kit · 7,104 tokens