cascadia CLAUDE.md

Project instructions for Cascadia, an agentic software template with an Anthropic SDK, tracing, and an orchestrator-and-subagent structure. They describe how coding agents should behave while working in the repository.

In plain words
What is it for?
Use them when making non-trivial decisions, changing the Cascadia project, or deciding how to report problems and scope choices.
Why use it?
They set expectations for concise communication, explicit assumptions, alternatives for significant decisions, and direct diagnosis of problems.

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/docraticmethod/cascadia/claude-md
Clone the repo
git clone --depth 1 https://github.com/docraticmethod/cascadia
Per session 1,154 This file is loaded in full into every session.
When invoked 1,154 The same file — it is already loaded in full.
Security scan A 1 finding. 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 $0.01154 $0.01154
Opus 5 $0.00577 $0.00577
Sonnet 5 $0.00231 $0.00231
Haiku 4.5 $0.00115 $0.00115

Measured 2d ago against content hash 8527426182c3, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

cascadia CLAUDE.md scanned grade A 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 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.

Makes network callslowCapability

Not a fault in itself. Listed so you know the mod talks to something, and to what.

curl -s http://localhost:3000/health
CLAUDE.md · 110 lines

How it starts

The opening of the file, as written. The whole thing — 110 lines — stays where its author put it; the contents beside it link to each section on GitHub.

CLAUDE.md

Opinionated agentic template.

Includes Anthropic SDK, traces directory, eval pattern, the orchestrator/subagent/skills shape. Reuse target is "another agentic harness".

This file provides guidance to Claude Code (claude.ai/code) when working with code in this repository.

Behavioral Instructions

  • Be concise. No preamble like "Great question" or "I'll help you with that."
  • State conclusions first, then reasoning if asked. Don't bury the answer.
  • Push back on bad ideas with reasoning, not deference.
  • Skip apologies for normal corrections.

Project Directives

  • For non-trivial decisions, surface 2–3 options with tradeoffs (cost, complexity, scalability) before recommending one. Skip if the choice is obvious.
  • Tag each significant decision with its business rationale: who benefits, what outcome it supports.
  • When something isn't working, name it explicitly. Don't paper over failures — diagnose, adjust, learn.
  • Note assumptions inline. Mark anything we're treating as given that could plausibly be wrong.
  • Flag scope cuts: when proposing a simpler version of something, say what we're not doing and what we'd do "with more time."
  • Prefer industry-standard patterns over clever ones. Document why if we deviate.

Scope

Work only within this directory. Do not read, modify, or reference files outside the repo root.

Working with the Agent

  • Plan before execute: propose changes before creating/modifying files for any non-trivial task
  • Prefer small, reviewable diffs over large sweeps
  • Surface assumptions explicitly; ask when ambiguous rather than guessing
  • After each change, summarize what was done and what's next

Project Overview

Express.js repo. The goal is to demonstrate clean, idiomatic Node.js API development with good engineering habits — small commits, tested code, clear decisions.

  • Runtime: Node 24
  • Module system: ESM ("type": "module" in package.json, .js extensions on all imports)
  • Framework: Express 5 (async errors propagate to error middleware automatically — no need to next(err))

Read the full file on GitHub · 110 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. 2d ago First seen · 110 lines · 1,154 tokens per session scan A 8527426182c3

Subscribe to this mod's changes

cascadia CLAUDE.md is an instructions file published in the GitHub repository docraticmethod/cascadia (11 stars, last pushed 16d ago), licensed Apache-2.0. It adds 1,154 tokens to every session, about $0.0058 per session on Opus 5. A static security scan graded it A with 1 finding (makes network calls). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.