What is adamas-lang CLAUDE.md?

Project instructions for working on the Adamas Crystal V2 compiler, including its language rules, documentation style, build commands, and safe test procedure. A compiler turns source code into a runnable program.

Use them when editing the compiler, writing project documentation, building it, running compiled test programs safely, or investigating crashes.

They reduce the risk of unsafe test runs, inconsistent project communication, accidental standard-library changes, and incorrect build or debugging steps.

adamas-lang CLAUDE.md — Instructions file for coding agents by skuznetsov: Project instructions for working on the Adamas Crystal V2…

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/skuznetsov/adamas-lang/claude-md

Clone the repo

git clone --depth 1 https://github.com/skuznetsov/adamas-lang

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/skuznetsov/adamas-lang/claude-md.svg)](https://agentmods.dev/instructions/skuznetsov/adamas-lang/claude-md)

Your own site

<a href="https://agentmods.dev/instructions/skuznetsov/adamas-lang/claude-md"><img src="https://agentmods.dev/badge/instructions/skuznetsov/adamas-lang/claude-md.svg" alt="Measured on agentmods" height="20"></a>

Per session 1,030 This file is loaded in full into every session.

When invoked 1,030 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.

Model	Per session	Once invoked
Fable 5.1	$0.01030	$0.01030
Opus 5	$0.00515	$0.00515
Sonnet 5	$0.00206	$0.00206
Haiku 4.5	$0.00103	$0.00103

Measured 5d ago against content hash 0376ee5073db, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-05, from the pricing page.

Security

Grade A, and why

adamas-lang 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 5d 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 · 57 lines

How it starts

The opening of the file, as written. The whole thing — 57 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Crystal V2 Compiler - Project Instructions

Language & documentation

Repository (code comments, markdown docs, commit messages in this project): write in English.
Chat with the project owner: prefer Russian unless they switch language.

Safe Testing Protocol

ALWAYS run test binaries via scripts/run_safe.sh <binary> <timeout> <max_mem_mb>
NEVER run test binaries directly — they can exhaust FDs/memory and freeze the machine
Default: scripts/run_safe.sh /tmp/test_hello 5 512 (5s timeout, 512MB limit)
The script monitors FD count and RSS; kills process if FD > 1000 or memory > limit

Build & Run

crystal build src/adamas.cr -o bin/adamas --error-trace   # build compiler
bin/adamas /tmp/test_hello.cr                                  # compile test
scripts/run_safe.sh /tmp/test_hello 5 512                          # run safely

Debug crash: lldb --batch -o 'run' -k 'bt 30' -k 'quit' /tmp/test_hello
Eager HIR (debug): ADAMAS_EAGER_HIR=1 bin/adamas ...

Critical Rules

NEVER modify stdlib files — must be 100% compatible with original Crystal stdlib at ../crystal/src
When in doubt, always check original Crystal compiler code at ../crystal/src
Original Crystal compiler: ../crystal/src/compiler/crystal/codegen/
Prefer zero-copy memory patterns to reduce GC pressure, bound memory use, and speed up the compiler.
Clean up temporary artifacts (/tmp/, ./tmp/, ~/Projects/Crystal/.codex_artifacts/) when no longer needed; commit working features and fixes when ready.
When you find a bug, add a regression script to reproduce it — it pays off quickly.
For fast debug iterations, prefer no-prelude oracles: easier to inspect .ll, .hir, and .mir.
Keep TODO.md updated when shipping meaningful changes.
Many bugs trace back to incorrect arena lifetime / tracking — verify arenas first.

Grok ACP Sidecar

Use Grok via ~/.grok/bin as a cheap read-only sidecar for non-trivial root-cause work.
Give Grok narrow bounded questions with exact files, logs, anchors, and required output shape; do not ask it to make edits.
Treat Grok output as candidate evidence only. Verify locally before adopting any claim.
Record useful and failed Grok interactions in grok_acp_beta_report.md so the beta-test signal survives context resets.
Prefer Grok for parallel audits such as "classify this missing-demand family", "find all call paths to this enqueue site", or "review this proposed root-cause hypothesis".
Do not wait on Grok when the next local falsifier is obvious; continue local work and integrate Grok only if it returns actionable evidence.

Read the full file on GitHub · 57 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.

5d ago First seen · 57 lines · 1,030 tokens per session scan A 0376ee5073db

Subscribe to this mod's changes

adamas-lang CLAUDE.md is an instructions file published in the GitHub repository skuznetsov/adamas-lang (52 stars, last pushed 4d ago), licensed MIT. It adds 1,030 tokens to every session, about $0.0052 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.

Languages Rust

turso AGENTS.md

AGENTS.md instructions for tursodatabase/turso, covering turso agent guidelines, quick reference, testing, running tests and test organization.

tursodatabase/turso · 2,194 tokens

cli AGENTS.md

AGENTS.md instructions for googleworkspace/cli, covering agents.md, project overview, build & test, changesets and architecture.

googleworkspace/cli · 2,700 tokens

azure-sdk-for-rust resourcemanager.instructions.md

Instructions for Azure/azure-sdk-for-rust, a project described as: This repository is for the active development of the Azure SDK for Rust. For consumers of the SDK we recommend visiting Docs.rs and looking up the docs for any of libraries in the SDK.

Azure/azure-sdk-for-rust · 400 tokens

adamas-lang CLAUDE.md