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.
npx agentmods add instructions/luumenlabs/code/agents-mdgit clone --depth 1 https://github.com/luumenlabs/codeWhat 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 | $0.00346 | $0.00346 |
| Opus 5 | $0.00173 | $0.00173 |
| Sonnet 5 | $0.00069 | $0.00069 |
| Haiku 4.5 | $0.00035 | $0.00035 |
Grade A, and why
code AGENTS.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 yesterday.
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.
Copies of this mod
1 near-identical copy found in the catalogue:
- code CLAUDE.md — 100% identical, 0 lines differ
What it actually says
Luu Code — notes for coding agents
House style
Comments are short. One or two lines. They explain why — the constraint that made the obvious approach wrong, the failure being defended against — and only where the next person would otherwise undo the code. Do not argue a case, do not compare the approach to the one not taken, do not narrate what the next line plainly does. A paragraph justifying a decision is a paragraph nobody reads. Match the density of the file you are editing.
User-facing text is shorter. Labels, descriptions, tooltips, and errors say what the thing is or what to do next. They never explain a reason, a trade-off, or how anything works underneath. "Always on — this is how an agent reports what is wrong" is already at the limit; anything longer belongs nowhere.
Documentation describes what is, not what was considered. No rationale for picking a library, no defence of an architecture, no history. Say how the thing behaves and what will break if you change it.
- Fail clearly. Every Roblox-facing failure carries a code an agent can act on and, where possible, a hint naming what to try instead.
- Never report a partial success as a success.
- Do not guess at a target. Ambiguity is an error, not a coin flip.
- Probe capabilities rather than assuming them; Studio differs by version, platform, and run state.
- Prefer a structured Studio API over desktop automation when both would work.
- Keep the surface small. The product is the Studio connection, not a general local coding environment.
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.
- yesterday First seen · 30 lines · 346 tokens per session scan A 8844afc3c530
code AGENTS.md is an instructions file published in the GitHub repository luumenlabs/code (2 stars, last pushed 16d ago), licensed MIT. It adds 346 tokens to every session, about $0.0017 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.
Other instructions, from other repositories
skills AGENTS.md
AGENTS.md instructions for ShiroKSH/skills, a project described as: Roblox Studio skill and local MCP bridge for Claude Code, Codex, and other coding agents.
raven-roblox-mcp AGENTS.md
AGENTS.md instructions for valrinx/raven-roblox-mcp, covering raven mcp engineering rules, evidence-first workflow and raven-specific invariants.
amp-acp AGENTS.md
Instructions for tao12345666333/amp-acp, covering agents.md, commands, architecture and code style.
roamcode AGENTS.md
Instructions for burakgon/roamcode, covering repository instructions for coding agents, public-repository safety, stable release and ota contract, releasing a stable version and ota changes.
dev-workspace AGENTS.md
Instructions for iwe-org/dev-workspace, covering agent operating manual, start of every session, the operating loop, conventions and iwe basics.
ob-1 AGENTS.md
Instructions for Overbrilliant/ob-1, covering agents.md, project index, stack, commands and key files.