What is z-spec CLAUDE.md?

A project guide for z-spec, a toolkit for writing and checking Z specifications, which are formal descriptions used to model and verify software behavior. It includes command-line tools, an MCP server, and a Claude Code plugin.

Use it when creating or checking specifications, running fuzzing or animation tools, changing command or surface code, or working within the Punt Labs workspace.

It explains required workspace setup and the boundary between commands and user-facing surfaces, helping agents follow the repository’s organization and verification rules.

How punt-labs/z-spec configures Claude Code

Reuse

Borrowing it

Nothing to install: this file belongs to punt-labs/z-spec. 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/punt-labs/z-spec/main/CLAUDE.md

Clone the repo

git clone --depth 1 https://github.com/punt-labs/z-spec

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/punt-labs/z-spec/claude-md/github.svg)](https://agentmods.dev/instructions/punt-labs/z-spec/claude-md)

Your own site

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

Per session 1,905 This file is loaded in full into every session.

When invoked 1,905 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	$0.01905	$0.01905
Opus 5.5	$0.00762	$0.00762
Sonnet 5.5	$0.00381	$0.00381
Haiku 4.5	$0.00191	$0.00191

Measured 3d ago against content hash 492581a56795, method: parsed. Prices are Anthropic first-party input rates as of 2026-10-07, from the pricing page.

Security

Grade A, and why

z-spec 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 3d 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 · 135 lines

How it starts

The opening of the file, as written. The whole thing — 135 lines — stays where its author put it; the contents beside it link to each section on GitHub.

z-spec

Formal Z specification toolkit: a fuzz/probcli wrapper, an MCP server (zspec), a CLI (z-spec), and a Claude Code plugin whose skill prompts guide spec creation, type-checking, and animation. Deterministic work — parsing, binary wrappers, report I/O, lux rendering — lives in the Python package; skills call its MCP tools instead of raw bash.

Part of Punt Labs. Must be checked out inside the punt-labs/ workspace meta-repo so org-wide configuration (git identity, beads DB, API keys) loads from ../CLAUDE.md and ../.envrc.

Package: punt-z-spec (PyPI)
CLI: z-spec
MCP server: zspec (stdio; z-spec mcp)
Python: 3.13+, managed with uv

Architecture

One engine, two thin client surfaces (CLI + MCP), both resolving the same Command objects from commands/registry.py — tests/commands/test_parity.py asserts they cannot drift. A command validates inputs, invokes the binary or store, and returns a CommandResult; it never prints and never knows which surface called it. display.py is the only module that publishes scenes to the lux Hub.

Build and test commands

make check      # full quality gate: lint, type, test, check-oo,
                 #   check-coupling, check-suppressions, check-dev-commands
make lint       # markdownlint + ruff check + ruff format --check + shellcheck
make type       # mypy + pyright + fuzz on every examples/*.tex spec
make test       # pytest + probcli model-check on every examples/*.tex spec
make uat        # build wheel, install CLI, then run the acceptance flight
                 #   by hand — RECONNECT the MCP server after; a reinstall
                 #   does not restart a running stdio server
make check-oo   # OO ratchet: must improve over baseline, never regress

make check passing means the code compiles, types hold, unit tests pass, and every spec type-checks and model-checks. It is necessary, never sufficient — it says nothing about what a person sees running the CLI, calling the MCP tool, or clicking the lux menu entry. The verification of record for every user-facing surface is the feature running in the installed artifact, exercised by hand against a written-in-advance expectation, before the PR opens.

Read the full file on GitHub · 135 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.

3d ago Changed · +2 lines · +14 tokens per session 492581a56795
11d ago Changed · +2 lines · +55 tokens per session 7924eae018be
2mo ago Changed · -312 lines · -4,181 tokens per session 50c6f2cedcaa
2mo ago First seen · 443 lines · 6,017 tokens per session scan A 9801cee2bba7

Subscribe to this mod's changes

z-spec CLAUDE.md is an instructions file published in the GitHub repository punt-labs/z-spec (5 stars, last pushed 2d ago), licensed MIT. It adds 1,905 tokens to every session, about $0.0076 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.

LaTeX beta claude-code-plugin formal-methods z-notation z-notations

z-spec CLAUDE.md next to 6 instructions from other repositories
Mod	Type	Stars	Last push	Licence	Scan	Tokens	Works with
z-spec CLAUDE.mdpunt-labs/z-spec	Instructions file	5	2d ago	MIT	A	1,905	Claude Code
vscode buildNext.instructions.mdmicrosoft/vscode	Instructions file	195k	today	MIT	A	9,719	GitHub Copilot
next.js AGENTS.mdvercel/next.js	Instructions file	143k	today	MIT	A	7,461	Codex, OpenCode
vscode oss-third-party-notices.instructions.mdmicrosoft/vscode	Instructions file	195k	today	MIT	A	5,516	GitHub Copilot
langchain AGENTS.mdlangchain-ai/langchain	Instructions file	149k	today	MIT	A	4,680	Codex, OpenCode
vscode copilot-instructions.mdmicrosoft/vscode	Instructions file	195k	today	MIT	A	3,093	GitHub Copilot
deepseek-harness AGENTS.mddeepseek-ai/deepseek-harness	Instructions file	275k	3d ago	MIT	A	4,313	Codex, OpenCode

z-spec: Instructions file for Claude Code