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/name0x0/hymarkx/agents-mdgit clone --depth 1 https://github.com/NAME0x0/HyMarkXWrote 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.
[](https://agentmods.dev/instructions/name0x0/hymarkx/agents-md)<a href="https://agentmods.dev/instructions/name0x0/hymarkx/agents-md"><img src="https://agentmods.dev/badge/instructions/name0x0/hymarkx/agents-md.svg" alt="Measured on agentmods" height="20"></a>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 | $0.00795 | $0.00795 |
| Opus 5 | $0.00398 | $0.00398 |
| Sonnet 5 | $0.00159 | $0.00159 |
| Haiku 4.5 | $0.00080 | $0.00080 |
Grade A, and why
HyMarkX 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 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.
How it starts
The opening of the file, as written. The whole thing — 70 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Agent instructions
For coding agents working in this repository: Codex, Claude Code, Cursor, Copilot, and anything else that reads a conventions file.
Read llms.txt first. It is the canonical description of the language and
lists the mistakes that are actually common. Everything below is about working on the
repository, not about writing .hmx documents.
Before changing anything
Read CLAUDE.md for the hard invariants. The ones that get violated most:
- A document with no HMX construct must render as CommonMark + GFM. The conformance suite (652 examples) is a CI gate and has never regressed across seven syntax additions. If your change moves that number, the change is wrong — do not adjust the suite.
- Only
@hymarkx/parsermay import micromark, mdast, hast, or unified. Enforced byscripts/check-boundaries.mjs, not by convention. - Only
@hymarkx/cliand@hymarkx/language-servermay touchnode:builtins. Everything else must run unchanged in a browser. - Every AST node carries a real source span. Dropping spans is a bug, not a shortcut.
- Recoverable problems are
Diagnostics with stable codes, never thrown exceptions. - Syntax changes need an ADR and a
SPEC.mdupdate before code.
Verification is not optional
pnpm check # build, lint, typecheck, test, boundary check
pnpm format:check
pnpm check runs pnpm build first, because several suites import workspace packages by
name and resolve through dist/. A green local run without building proves nothing.
Do not report success on the basis that code compiles. Every task in this project that reported success while carrying a real defect did so by trusting a step it had not run — a failing typecheck, a hidden workaround, a conformance regression, a verification using a library that was not installed.
Where decisions live
Nothing important lives in commit messages or chat. If you are about to make a judgement call, it is probably already recorded:
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 First seen · 70 lines · 795 tokens per session scan A 27e8ba3c8f32
HyMarkX AGENTS.md is an instructions file published in the GitHub repository NAME0x0/HyMarkX (2 stars, last pushed 9d ago), licensed Apache-2.0. It adds 795 tokens to every session, about $0.0040 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
hongdown AGENTS.md
AGENTS.md instructions for dahlia/hongdown, covering project architecture, initial setup, building, testing and website.
hongdown CLAUDE.md
Claude Code instructions for dahlia/hongdown, a project described as: A Markdown formatter that enforces Hong Minhee's Markdown style conventions.
goldmark AGENTS.md
AGENTS.md instructions for yuin/goldmark, covering instructions for agents, testing instructions, benchmark instructions, coding conventions and key point of commonmark spec.
markview CLAUDE.md
Instructions for paulhkang94/markview, covering markview — ai development guide, quick reference, architecture, mcp tools (for ai integration) and safety rules.
mdscript CLAUDE.md
Instructions for dean0x/mdscript, covering mds (markdown script), build and test, release, land the bump on main via pr (ci-gated), then and gotchas.
dingo CLAUDE.md
Claude Code instructions for MadAppGang/dingo, covering claude ai agent instructions - dingo project, 🚨🚨🚨 stop: read this before any implementation 🚨🚨🚨, the architectural principle (understand this first), why this rule exists and ✅ required approaches (the only correct ways).