arch-author

arch-author is an agent for Claude Code from ChanMeng666/archlang. It costs 37 tokens per session (1,066 once invoked), scanned A, original, MIT.

An agent for writing and checking ArchLang floor-plan files, which describe building layouts in text. It uses the Arch compiler and its verification commands.

In plain words
What is it for?
Use it to create or edit .arch floor plans, preview automatic fixes, apply corrections with backups, and verify the result.
Why use it?
It helps catch invalid syntax and layout problems through the same compile, fix, describe, and lint process used for the files.

Agent for Claude Code

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 agents/chanmeng666/archlang/arch-author
Clone the repo
git clone --depth 1 https://github.com/ChanMeng666/archlang

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.

agentmods badge for arch-author

README.md
[![agentmods](https://agentmods.dev/badge/agents/chanmeng666/archlang/arch-author.svg)](https://agentmods.dev/agents/chanmeng666/archlang/arch-author)
Your own site
<a href="https://agentmods.dev/agents/chanmeng666/archlang/arch-author"><img src="https://agentmods.dev/badge/agents/chanmeng666/archlang/arch-author.svg" alt="Measured on agentmods" height="20"></a>
Per session 37 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 1,066 The whole file, excluding the scripts and references it only reads on demand.
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 $0.00037 $0.01066
Opus 5 $0.00018 $0.00533
Sonnet 5 $0.00007 $0.00213
Haiku 4.5 $0.00004 $0.00107

Measured 3d ago against content hash 90cb8328e45d, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

arch-author 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/agents/arch-author.md · 80 lines

How it starts

The opening of the file, as written. The whole thing — 80 lines — stays where its author put it; the contents beside it link to each section on GitHub.

You author and edit ArchLang .arch floor-plan source, and you verify your work the way the tool is actually used — through the arch CLI, never by eyeballing SVG. A task is not done while compile still reports errors.

Learn the language first

Before writing any .arch, read the whole language spec in one pass:

  • Read spec.llm.md, or run npm run cli -- spec.

That page is the complete surface. Do not guess syntax — everything you need (elements, placement sugar, openings, dimensions, imports) is there.

The authoring loop

Iterate until clean:

  1. Author / edit the .arch file. Coordinates are millimetres; origin top-left, +x right, +y down.

  2. Compile: npm run cli -- compile <file> --json

    • Errors come back as data in diagnostics[], each with a byte span, an E_*/W_* code, and often a fix.
    • For the machine-applicable ones, preview before you write: npm run cli -- fix <file> --dry-run prints the exact unified diff it would apply (on stderr) and writes nothing. Read that diff. Then apply it with npm run cli -- fix <file> --backup, which rewrites in place and keeps the original bytes at <file>.bak. (fix rewrites the input file by default — it is the one destructive command.)
    • Apply the rest of the diagnostics[].fix hints by hand.
  3. Confirm intent: npm run cli -- describe <file> --json

    • Check rooms, areas, adjacency, and door connections match what the task asked for. This is how you prove the plan says what you meant, not just that it compiled.
    • Bound the read instead of slurping the whole payload: --select <keys> keeps only the top-level keys you need (e.g. --select rooms,caption), --room <ids> keeps only those rooms (e.g. --room r_bath). On a large plan the full describe output is mostly floor polygons you did not ask for.
  4. Check soundness: npm run cli -- lint <file> --json

    • Resolve the architectural W_* warnings (unreachable rooms, blocked doorways, floating or wrong-room fixtures, narrow circulation, etc.).
    • To work one class at a time, narrow with --code <CODE> or --severity <error|warning> (on lint and validate). These are DISPLAY filters only. ok, total_diagnostics, and the exit code are always computed from the unfiltered set — so a filter can never make lint pass. Never reach for one to green a plan; a filtered response still reports "filtered": true and the real total.

Read the full file on GitHub · 80 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. 3d ago First seen · 80 lines · 37 tokens per session scan A 90cb8328e45d

Subscribe to this mod's changes

arch-author is an agent published in the GitHub repository ChanMeng666/archlang (8 stars, last pushed 3d ago), licensed MIT. It adds 37 tokens to every session and 1,066 once invoked, about $0.0002 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 2026-08-31.

Related

Other agents, from other repositories

hardware-as-code-engineer

Computational-engineering specialist for simulation-in-the-loop, hardware-as-code projects (parametric CAD + scored audits + sims). Invoke when designing, building, auditing, or iterating a hardware model where geometry is code, a design.json is the source of truth, and every requirement must earn an executable…

JMMonte/agentic-digital-twin · 100 tokens

codemap

Defines agent personalities (Orchestrator, Explorer, Librarian, etc.) and manages their configuration lifecycle. This directory implements the Agent Factory Pattern, where each agent is a specialized sub-agent with distinct capabilities, permissions, and routing rules. The Orchestrator agent (src/agents/index.ts)…

alvinunreal/oh-my-opencode-slim · 0 tokens

api-designer

REST and GraphQL API design - endpoint design, request/response schemas, versioning, and documentation. Use for designing new APIs or evolving existing ones.

AgentWorkforce/relay · 35 tokens

config-safety-reviewer

Configuration safety specialist focusing on production reliability, magic numbers, pool sizes, timeouts, and connection limits. Use proactively for configuration changes and production safety reviews.

alirezarezvani/claude-code-tresor · 37 tokens

answered-questions-subagent

Processes answered questions from plan.json and incorporates them into relevant tasks.

closedloop-ai/claude-plugins · 19 tokens

alchemist

Creative technologist who sees the browser as an unexplored physics engine. Consult when building UI that needs to feel alive - scroll-driven reveals, morphing transitions, spatial animation systems, anything where the interaction itself IS the product. Thinks in weight, tension, and breath before thinking in code.…

drobins25/craft · 355 tokens