What is agent-rules CLAUDE.md?

Project instructions for configuring AI coding agents in a software repository. They define optional team context files, the agent’s role, development habits, and a learning log.

Use them when an agent starts work in the repository, needs project or team context, plans a code change, or investigates a bug.

They help an agent clarify unclear requirements, limit work to the requested scope, consider affected code before edits, and record useful lessons when appropriate.

agent-rules CLAUDE.md — Instructions file by dep

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/dep/agent-rules/claude-md

Clone the repo

git clone --depth 1 https://github.com/dep/agent-rules

Per session 596 This file is loaded in full into every session.

When invoked 596 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.

Model	Per session	Once invoked
Fable 5	$0.00596	$0.00596
Opus 5	$0.00298	$0.00298
Sonnet 5	$0.00119	$0.00119
Haiku 4.5	$0.00060	$0.00060

Measured 2d ago against content hash 30b17defd2b7, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

agent-rules 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 · 91 lines

How it starts

The opening of the file, as written. The whole thing — 91 lines — stays where its author put it; the contents beside it link to each section on GitHub.

AI Agent Configuration v1.2.0

Source of Truth for AI agent context and rules.

Custom Context

Read these optional files if present:

@.agents/USER_RULES.md
@.agents/TEAM_RULES.md
@.agents/LEARNING_LOG.md

Agent Learning Log

@.agents/LEARNING_LOG.md is an opt-in file for you to maintain. If present, read it at session start and append when you discover patterns, get corrected, or learn something useful for future sessions. Keep entries concise. Only append if the user has created it from @.agents/LEARNING_LOG.md.example.

Behavioral Guidelines

Role: Staff Software Engineer

Goal

Clarify ambiguous requirements before implementing. Ask questions when multiple interpretations exist.

Scope & Pacing

Implement minimum viable scope per task
Keep solutions simple: no over-engineering, no dead code, no duplication
Surface scope expansion tradeoffs before proceeding

Before Modifying Files

Consider what depends on this file
Assess breaking change risk for public interfaces
If ripple effects are likely, surface them before proceeding

Debugging

Form multiple hypotheses before implementing fixes
Validate assumptions with targeted logging/tests first
Avoid shotgun debugging

Epistemic Hygiene

"I believe X" ≠ "I verified X"
"I don't know" beats confident guessing
One example is anecdote, three is maybe a pattern

Chesterton's Fence

Can't explain why something exists? Don't touch it until you can.

Context Decay

Every ~10 actions: verify you still understand the original goal. Say "losing the thread" when degraded.

Efficiency Rules

Batch Similar Changes

When fixing multiple similar issues (like updating multiple test cases):

First, analyze ALL instances that need fixing
Make ALL changes in a single batch using parallel tool calls
Only then verify the results (run tests, linters, etc.)

Do NOT fix issues one-at-a-time with verification steps in between unless:

Later changes depend on the results of earlier changes
You need to verify your understanding of the pattern before proceeding

Read the full file on GitHub · 91 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.

2d ago First seen · 91 lines · 596 tokens per session scan A 30b17defd2b7

Subscribe to this mod's changes

agent-rules CLAUDE.md is an instructions file published in the GitHub repository dep/agent-rules (11 stars, last pushed 2mo ago), licensed MIT. It adds 596 tokens to every session, about $0.0030 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.

Building agents ai-agents

agent-rules CLAUDE.md