What is wanaku CLAUDE.md?

A Rust-based proxy that sits between AI agents and backend systems. It checks tool calls, messages between agents, and model traffic so agents do not connect to backends directly.

Use it when building or running an agent system that needs controlled access to backend tools, agent-to-agent communication, or model services through configurable pipelines.

It centralises policy, identity, data-control, and audit checks in one intermediary layer. This gives a project one place to control and record agent interactions with backend services.

How wanaku-ai/wanaku configures Claude Code

Reuse

Borrowing it

Nothing to install: this file belongs to wanaku-ai/wanaku. 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/wanaku-ai/wanaku/main/CLAUDE.md

Clone the repo

git clone --depth 1 https://github.com/wanaku-ai/wanaku

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.

README.md

[![agentmods](https://agentmods.dev/badge/instructions/wanaku-ai/wanaku/claude-md/github.svg)](https://agentmods.dev/instructions/wanaku-ai/wanaku/claude-md)

Your own site

<a href="https://agentmods.dev/instructions/wanaku-ai/wanaku/claude-md"><img src="https://agentmods.dev/badge/instructions/wanaku-ai/wanaku/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.

Your own site · 80×15

<a href="https://agentmods.dev/instructions/wanaku-ai/wanaku/claude-md"><img src="https://agentmods.dev/badge/instructions/wanaku-ai/wanaku/claude-md.svg" alt="Reviewed on agentmods" width="80" height="20"></a>

Per session 3,212 This file is loaded in full into every session.

When invoked 3,212 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.

Model	Per session	Once invoked
Fable 5.1	$0.03212	$0.03212
Opus 5.5	$0.01285	$0.01285
Sonnet 5.5	$0.00642	$0.00642
Haiku 4.5	$0.00321	$0.00321

Measured 6d ago against content hash 2925b7720d6c, method: parsed. Prices are Anthropic first-party input rates as of 2026-10-07, from the pricing page.

Security

Grade A, and why

wanaku 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.

CLAUDE.md · 252 lines

How it starts

The opening of the file, as written. The whole thing — 252 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Wanaku

Governed action proxy for AI agents, built in Rust on the Praxis proxy framework. Sits between agents and backend systems, intercepting tool calls, agent-to-agent messages, and inference traffic. Policy, identity, data controls, and audit are enforced in the proxy layer — agents never touch backends directly.

Core Guidelines

Think before you write. Don't create abstractions unnecessarily. Simplicity matters — focus on minimum code to achieve the result. Don't cheat lint or clippy rules.

Quick Start

cargo build && cargo test
cargo run  # MCP :8081, management API :8080
cargo run -- --pipeline-config pipeline.yaml --wanaku-config wanaku.yaml  # custom configs

Default configs: server/src/default.yaml (embedded, pipeline), wanaku.yaml (optional, forward registry bootstrap).

Critical: Changes to server/src/default.yaml don't trigger rebuilds (include_str!). Touch server/src/lib.rs to force recompile.

Config modification at startup: load_config in server/src/lib.rs parses the embedded default.yaml into serde_yaml::Value and applies env var overrides programmatically (inference upstream, TLS SNI, CORS origin) before passing to praxis. Helper functions find_named_entry_mut, find_inference_cluster, apply_inference_config, and apply_cors_config navigate the YAML tree. Falls back to raw defaults with error logging if parsing or serialization fails.

Build Info

Rust 2024, MSRV 1.96
Lints: #![deny(unsafe_code, unwrap_used, expect_used, panic)] in all crates
Jemalloc on Unix platforms

Architecture

Workspace: types/ (shared types, traits, Feature trait, context structs — zero heavy deps), infra/ (infra: LLM client, MCP client, metrics, persistence, registry impl), filters/ (core MCP filters), features/ (self-contained feature crates: mcp-metadata, evaluator, intercept, chat), server/ (binary, pipeline, mgmt API), ui/admin/ (React embedded UI).

Feature crates that only need Feature/HttpContext/registry traits depend on wanaku-types; crates needing concrete infra (LLM calls, MCP forwarding, metrics counters, file persistence) also depend on wanaku-infra. See docs/get-started/architecture.md for the full boundary.

Read the full file on GitHub · 252 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.

6d ago Changed · +3 lines · +296 tokens per session 2925b7720d6c
8d ago Changed · +33 tokens per session 6656bdbb3726
9d ago Changed · +13 tokens per session 14b895218d92
19d ago Changed · +21 tokens per session db2561d1951e
24d ago Changed · +12 lines 075ce56cf42b
2mo ago Changed · +7 lines · +104 tokens per session da3df1bf18c3
2mo ago First seen · 230 lines · 2,745 tokens per session scan A 14fa7583e4af

Subscribe to this mod's changes

wanaku CLAUDE.md is an instructions file published in the GitHub repository wanaku-ai/wanaku (135 stars, last pushed yesterday), licensed Apache-2.0. It adds 3,212 tokens to every session, about $0.0128 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.

Backend and APIs Building agents agentic agentic-ai agents artificial-intelligence mcp mcp-server

wanaku CLAUDE.md next to 6 instructions from other repositories
Mod	Type	Stars	Last push	Licence	Scan	Tokens	Works with
wanaku CLAUDE.mdwanaku-ai/wanaku	Instructions file	135	yesterday	Apache-2.0	A	3,212	Claude Code
vscode buildNext.instructions.mdmicrosoft/vscode	Instructions file	195k	yesterday	MIT	A	9,719	GitHub Copilot
next.js AGENTS.mdvercel/next.js	Instructions file	143k	today	MIT	A	7,461	Codex, OpenCode
langchain AGENTS.mdlangchain-ai/langchain	Instructions file	149k	yesterday	MIT	A	4,680	Codex, OpenCode
vscode oss-third-party-notices.instructions.mdmicrosoft/vscode	Instructions file	195k	yesterday	MIT	A	5,516	GitHub Copilot
vscode copilot-instructions.mdmicrosoft/vscode	Instructions file	195k	yesterday	MIT	A	3,093	GitHub Copilot
deepseek-harness AGENTS.mddeepseek-ai/deepseek-harness	Instructions file	274k	2d ago	MIT	A	4,313	Codex, OpenCode

wanaku: Instructions file for Claude Code