zapcode CLAUDE.md

Project-specific instructions for an AI coding assistant working on Zapcode, a project that parses TypeScript, compiles it to bytecode, and runs it in a virtual machine. It also defines where key parts of the codebase are located and which safety rules to follow.

In plain words
What is it for?
Use it to orient work on parsing, the intermediate representation, compilation, bytecode, the virtual machine, and snapshots. It also sets a required reading order and a Git commit rule.
Why use it?
It gives the assistant the project’s architecture and security constraints before it changes code, reducing the risk of violating important sandbox rules.

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/theuncharted/zapcode/claude-md
Clone the repo
git clone --depth 1 https://github.com/TheUncharted/zapcode
Per session 1,620 This file is loaded in full into every session.
When invoked 1,620 The same file — it is already loaded in full.
Security scan A 0 findings. 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 $0.01620 $0.01620
Opus 5 $0.00810 $0.00810
Sonnet 5 $0.00324 $0.00324
Haiku 4.5 $0.00162 $0.00162

Measured 2d ago against content hash 23d109e202e5, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

zapcode 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 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.

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 · 188 lines

How it starts

The opening of the file, as written. The whole thing — 188 lines — stays where its author put it; the contents beside it link to each section on GitHub.

CLAUDE.md

AI coding assistant instructions for the zapcode project. See AGENTS.md for the full architecture reference. This file adds Claude-Code-specific guidance on top of it.


Git

Do not add Co-Authored-By trailers to git commits.


Read AGENTS.md first

Before writing any code in this repository, read AGENTS.md in full. It defines:

  • What Zapcode is and what it must never do
  • The full architecture (parser → IR → bytecode → VM → snapshot)
  • The five sandbox invariants that must never be violated
  • The definition of "done" for any feature

Do not skip this. The sandbox invariants in particular will save you from introducing security vulnerabilities that are hard to detect and easy to ship.


Codebase orientation

Start here when working on a new area:

Area Entry point
Parsing TypeScript crates/zapcode-core/src/parser/mod.rs
IR definition crates/zapcode-core/src/parser/ir.rs
Bytecode instructions crates/zapcode-core/src/compiler/instruction.rs
Compiler (IR → bytecode) crates/zapcode-core/src/compiler/mod.rs
VM main loop + dispatch crates/zapcode-core/src/vm/mod.rs
Built-in functions crates/zapcode-core/src/vm/builtins.rs
Value / type system crates/zapcode-core/src/value.rs
Snapshot / resume crates/zapcode-core/src/snapshot.rs
Resource limits crates/zapcode-core/src/sandbox.rs
Error types crates/zapcode-core/src/error.rs
JS bindings API crates/zapcode-js/src/lib.rs
Python bindings API crates/zapcode-py/src/lib.rs
WASM bindings API crates/zapcode-wasm/src/lib.rs

When in doubt about where something belongs: zapcode-core is pure Rust with zero I/O. Bindings crates only translate types and marshal calls into zapcode-core. Never put business logic in binding crates.


How to add a new language feature

  1. Check the supported subset table in AGENTS.md first. If the feature is explicitly listed as unsupported, do not add it without opening a discussion. Features are excluded intentionally.

Read the full file on GitHub · 188 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 · 188 lines · 1,620 tokens per session scan A 23d109e202e5

Subscribe to this mod's changes

zapcode CLAUDE.md is an instructions file published in the GitHub repository TheUncharted/zapcode (89 stars, last pushed 3d ago), licensed MIT. It adds 1,620 tokens to every session, about $0.0081 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.