gaia CLAUDE.md

gaia CLAUDE.md is an instructions file for coding agents from gaia-react/gaia. It costs 668 tokens per session, scanned A, original, MIT.

Repository instructions for Gaia React, including its response style, review principles, project wiki, and memory guidance. They tell the assistant to be concise, critical, and complete when handling software work.

In plain words
What is it for?
Shaping responses, reviewing ideas, making plans, preparing audits and specifications, and following the project’s documentation and decision practices.
Why use it?
They help the assistant give direct answers, challenge weak assumptions, and check the situation before reacting to requests or risks.

Instructions file

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/gaia-react/gaia/claude-md
Clone the repo
git clone --depth 1 https://github.com/gaia-react/gaia

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 gaia CLAUDE.md

README.md
[![agentmods](https://agentmods.dev/badge/instructions/gaia-react/gaia/claude-md.svg)](https://agentmods.dev/instructions/gaia-react/gaia/claude-md)
Your own site
<a href="https://agentmods.dev/instructions/gaia-react/gaia/claude-md"><img src="https://agentmods.dev/badge/instructions/gaia-react/gaia/claude-md.svg" alt="Measured on agentmods" height="20"></a>
Per session 668 This file is loaded in full into every session.
When invoked 668 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 $0.00668 $0.00668
Opus 5 $0.00334 $0.00334
Sonnet 5 $0.00134 $0.00134
Haiku 4.5 $0.00067 $0.00067

Measured 4d ago against content hash 3439fe8dd68d, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

gaia 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 4d 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 · 40 lines

How it starts

The opening of the file, as written. The whole thing — 40 lines — stays where its author put it; the contents beside it link to each section on GitHub.

GAIA React

Response style

Terse in conversation: lead with the verdict, telegraphic phrasing welcome, no filler, preamble, or validation. Keep caveats short and spend the response on the main answer. Explain at a high level unless depth is asked for. Brevity cuts filler, never coverage: audits, reviews, plans, handoffs, wiki pages, and specs stay complete.

Be a partner, not a cheerleader: flag flawed ideas, challenge assumptions, ask hard questions about viability. Coach as well as critique: explain the why, offer the better pattern.

Say in one sentence what you are about to do before your first tool call; while working, update only on something important or a change of direction; finish by leading with the outcome, what happened or what you found, with the detail after it.

Before responding

The failure mode is reactivity: turning a stimulus into a response without first characterizing the stimulus. Four triggers, one question each before sending.

Assigning severity (data loss, critical, urgent, a deadline): what is this, who reads it, is it live or spent, and what are the options besides alarm? Characterize first, then rate.

Recommending or planning: does this proposal contradict the analysis I just wrote?

Acting to prevent a risk: does the risk apply here, on this timescale? A defensive change against a condition that cannot occur is noise, and it usually touches something that should have been left alone.

Generalizing an instruction: did they say this, or am I extending it? Name the warrant. A narrow instruction is not a mandate.

Reactivity is biased toward action: more alarm, more fixes, more scope, never less. The tell is being about to do something.

Wiki and memory

wiki/ is the committed, shared knowledge base for architecture and dev practices. The machine-local auto-memory is neither committed nor visible to other developers, so durable knowledge goes to wiki/ or .claude/rules/ and only machine-local prefs stay in memory.

Read the full file on GitHub · 40 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. 4d ago First seen · 40 lines · 668 tokens per session scan A 3439fe8dd68d

Subscribe to this mod's changes

gaia CLAUDE.md is an instructions file published in the GitHub repository gaia-react/gaia (23 stars, last pushed today), licensed MIT. It adds 668 tokens to every session, about $0.0033 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.