What is decompose CLAUDE.md?

Project instructions for Decompose, a Python system that breaks text into structured parts such as risks, authorities, dates, entities, and content types without using an AI language model.

It guides work on the text-splitting, classification, entity-extraction, preservation, library, and command-line interfaces.

It helps keep important meaning, limits, and references organized before another system processes the text.

How echology-io/decompose configures Claude Code

Reuse

Borrowing it

Nothing to install: this file belongs to echology-io/decompose. 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/echology-io/decompose/main/CLAUDE.md

Clone the repo

git clone --depth 1 https://github.com/echology-io/decompose

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/echology-io/decompose/claude-md/github.svg)](https://agentmods.dev/instructions/echology-io/decompose/claude-md)

Your own site

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

Per session 689 This file is loaded in full into every session.

When invoked 689 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	$0.00689	$0.00689
Opus 5.5	$0.00276	$0.00276
Sonnet 5.5	$0.00138	$0.00138
Haiku 4.5	$0.00069	$0.00069

Measured 2mo ago against content hash 8d633dc431bf, method: parsed. Prices are Anthropic first-party input rates as of 2026-10-07, from the pricing page.

Security

Grade A, and why

decompose 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 2mo 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 · 61 lines

How it starts

The opening of the file, as written. The whole thing — 61 lines — stays where its author put it; the contents beside it link to each section on GitHub.

decompose — Project Instructions

Inherits from ../CLAUDE.md. Read that first.

What This Is

Decompose is the deterministic foundation of the echology system. It classifies any text into structured semantic units — authority, risk, attention, entities — without an LLM. No probability. No hallucination. No cost.

This is the most important project in echology. Every other system depends on it. Changes here propagate everywhere.

Architecture

Five modules, one pipeline:

text → chunker → classifier → entities → irreducibility → DecomposeResult

Module	What It Does
`chunker.py`	Semantic chunking, Markdown-aware, sentence-boundary splitting
`classifier.py`	Regex-based authority + risk + content type classification. Attention scoring.
`entities.py`	Regex entity extraction: standards, dates, financial values, legal references
`irreducibility.py`	Detects content that must be preserved verbatim (specs, limits, formulas)
`core.py`	Orchestration. `decompose_text()` and `filter_for_llm()`

Interfaces

Python library: from decompose import decompose_text, filter_for_llm
CLI: decompose --text "..." --pretty or piped stdin
MCP server: decompose-mcp --serve (exposes decompose_text and decompose_url tools)

Rules

Zero runtime dependencies. The library itself imports nothing outside the standard library. mcp is only required for MCP server mode. Do not add dependencies.
Deterministic. Same input always produces same output. No randomness. No LLM calls. No network calls (except decompose_url which fetches the URL, then classifies deterministically).
Published on PyPI as decompose-mcp. Changes must not break the public API (decompose_text, filter_for_llm, DecomposeResult, Unit).
63 tests. Run pytest before any commit to a core module. Do not reduce test coverage.
Regex patterns are the core IP. When modifying classifier patterns, test against real documents from multiple domains (AEC, insurance, legal, general). A pattern that improves one domain but breaks another is rejected.

Read the full file on GitHub · 61 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.

2mo ago First seen · 61 lines · 689 tokens per session scan A 8d633dc431bf

Subscribe to this mod's changes

decompose CLAUDE.md is an instructions file published in the GitHub repository echology-io/decompose (10 stars, last pushed 5mo ago), licensed MIT. It adds 689 tokens to every session, about $0.0028 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 2026-08-31.

Deterministic ai-agents classification document-intelligence mcp nlp

decompose CLAUDE.md next to 6 instructions from other repositories
Mod	Type	Stars	Last push	Licence	Scan	Tokens	Works with
decompose CLAUDE.mdechology-io/decompose	Instructions file	10	5mo ago	MIT	A	689	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	yesterday	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

decompose: Instructions file for Claude Code