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.
npx agentmods add instructions/mericcintosun/avakit/claude-mdgit clone --depth 1 https://github.com/mericcintosun/AvaKitWrote 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.
[](https://agentmods.dev/instructions/mericcintosun/avakit/claude-md)<a href="https://agentmods.dev/instructions/mericcintosun/avakit/claude-md"><img src="https://agentmods.dev/badge/instructions/mericcintosun/avakit/claude-md.svg" alt="Measured on agentmods" height="20"></a>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 5.1 | $0.00958 | $0.00958 |
| Opus 5 | $0.00479 | $0.00479 |
| Sonnet 5 | $0.00192 | $0.00192 |
| Haiku 4.5 | $0.00096 | $0.00096 |
Grade A, and why
AvaKit 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 6d 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.
How it starts
The opening of the file, as written. The whole thing — 44 lines — stays where its author put it; the contents beside it link to each section on GitHub.
CLAUDE.md — AvaKit
Guidance for AI agents (Claude Code / Cursor) working in this repository.
What this is
AvaKit — an open-source, AI-native Avalanche developer toolkit: one core with surfaces (@avakit/core, @avakit/react, create-avalanche-app, @avakit/mcp, plus the @avakit/studio local dashboard). Positioning: "Avalanche's open-source, AI-native create-next-app." Strategy: wrap mature pieces, don't rewrite them. See README.md and docs/00–docs/11.
Current state: 0.x pre-release, shipped. All five packages are published on npm and proven live on Fuji; 8 templates ship. APIs may still change before 1.0. (The docs/00–docs/10 planning specs predate the build and are partly stale — treat the code, README.md, and the website docs as the source of truth.)
Binding rules (do not violate)
These come from the project owner and override convenience. Full text in docs/11-conventions.md.
- Language: Everything written into the repo is English (code, comments, commits, docs). Conversation with the owner is Turkish, but no Turkish goes into project files.
- UI library: shadcn/ui ONLY. No other component library — and not BuilderKit's UI. Avalanche-specific components are built on shadcn primitives.
- Animation: Framer Motion or GSAP only.
- Color: The "black & white until M3" rule is lifted (owner, 2026-07). The website (
apps/web) now uses a brand color — "Ember Crimson" (a distinct crimson that evokes Avalanche's red/dark energy without copying it). Surfaces stay near-black/white; crimson is an accent (primary buttons, focus rings, links, eyebrow ticks,$/checkmarks). Dark/light wired from day one (next-themes), both fully supported. Colors live only inapps/web/app/globals.cssdesign tokens — components never hardcode them. Keep the scaffolder templates visually neutral/grayscale (users brand their own apps) unless the owner says otherwise. - Design: 2026-modern, professional devtools aesthetic (Linear / Vercel / shadcn-dashboard polish).
- Versions: Latest stable of every frontend tech (e.g. Next.js 16, React 19, Tailwind v4). Re-check "latest stable" at implementation time.
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.
- 6d ago First seen · 44 lines · 958 tokens per session scan A 43d3358bfb21
AvaKit CLAUDE.md is an instructions file published in the GitHub repository mericcintosun/AvaKit (12 stars, last pushed 3d ago), licensed MIT. It adds 958 tokens to every session, about $0.0048 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.
Other instructions, from other repositories
metamask-mobile AGENTS.md
AGENTS.md instructions for MetaMask/metamask-mobile, covering agents.md, project overview, essential commands, setup & installation and expo development (recommended for js-only development).
solana-com AGENTS.md
AGENTS.md instructions for solana-foundation/solana-com, covering solana.com monorepo agent guide, what this repo is, fast routing map, repo layout and apps.
solana-com CLAUDE.md
Claude Code instructions for solana-foundation/solana-com, covering solana.com monorepo, monorepo structure, tech stack, development commands and install dependencies.
metamask-mobile CLAUDE.md
Claude Code instructions for MetaMask/metamask-mobile, a project described as: Mobile web browser providing access to websites that use the Ethereum blockchain.
aptos-ts-sdk AGENTS.md
AGENTS.md instructions for aptos-labs/aptos-ts-sdk, covering agents.md, project overview, runtime compatibility, requirements and repo layout.
appwrite AGENTS.md
AGENTS.md instructions for appwrite/appwrite, covering appwrite, commands, stack, layout and libraries.