agentmods.dev

mcp-htmleditor AGENTS.md

mcp-htmleditor AGENTS.md is an instructions file for Codex, OpenCode from smorand/mcp-htmleditor. It costs 1,640 tokens per session, scanned A, original, MIT.

A set of coding-agent instructions for mcp-htmleditor, a visual HTML editor and MCP server with PowerPoint and Word export. It documents project commands, templates, testing, and development conventions.

In plain words
What is it for?
Use it when developing, testing, formatting, checking, installing, or running mcp-htmleditor.
Why use it?
It gives an agent the repository-specific workflow and checks needed to change the project safely and consistently.

Instructions file for CodexOpenCode

not rated 0 11d ago A scan 1,640 tokens original MIT
View the source file smorand/mcp-htmleditor
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/smorand/mcp-htmleditor/agents-md
Clone the repo
git clone --depth 1 https://github.com/smorand/mcp-htmleditor

Made for: Codex, OpenCode.

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 mcp-htmleditor AGENTS.md

README.md
[![agentmods](https://agentmods.dev/badge/instructions/smorand/mcp-htmleditor/agents-md.svg)](https://agentmods.dev/instructions/smorand/mcp-htmleditor/agents-md)
Your own site
<a href="https://agentmods.dev/instructions/smorand/mcp-htmleditor/agents-md"><img src="https://agentmods.dev/badge/instructions/smorand/mcp-htmleditor/agents-md.svg" alt="Measured on agentmods" height="20"></a>
Per session 1,640 This file is loaded in full into every session.
When invoked 1,640 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 $0.01640 $0.01640
Opus 5 $0.00820 $0.00820
Sonnet 5 $0.00328 $0.00328
Haiku 4.5 $0.00164 $0.00164

Measured 3d ago against content hash 6d0b96377c82, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

mcp-htmleditor 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.

AGENTS.md · 107 lines

How it starts

The opening of the file, as written. The whole thing — 107 lines — stays where its author put it; the contents beside it link to each section on GitHub.

AGENTS.md, mcp-htmleditor

Overview

WYSIWYG HTML editor plus MCP server, with PPTX and DOCX export. LLM agents modify HTML files on disk; the browser renders them in an iframe and reloads on mtime change. Python 3.13+, uv, click, FastMCP, python-pptx, BeautifulSoup4, pandoc for DOCX.

Key commands

make is the only entry point. Never call uv, ruff, mypy or pytest directly.

make sync        # uv sync (deps + dev group)
make check       # GATE: lint + format-check + typecheck + security + test-cov (80 %)
make test        # pytest -v            (ARGS='-k pptx' for a subset)
make install     # uv tool install + templates + log dir + Pi skill
make run ARGS='export pptx in.html out.pptx'
make docker-build / run-up / run-down
make bootstrap-ei  # regenerate the EI bootstrap from the EI reference
make sync-medical-css  # propagate the medical charter CSS to the reference decks
make help        # every target

CLI (same after make install):

mcp-htmleditor --version
mcp-htmleditor templates                   # ei, carbon, medical, doc, doc-perso, doc-ei, mail, website
mcp-htmleditor new ei pres.html --serve
mcp-htmleditor new medical talk.html --serve
mcp-htmleditor serve file.html [-v|-q] [--host] [--port] [--poll] [--no-browser]
mcp-htmleditor mcp                         # MCP server (stdio)
mcp-htmleditor export pptx in.html out.pptx
mcp-htmleditor export docx in.html out.docx
mcp-htmleditor arch-layout in.html         # compute arch-diagram positions from topology
mcp-htmleditor arch-checklist               # print the editable arch-diagram QA checklist
mcp-htmleditor skill                       # full skill content

Essential conventions

  • LLM workflow: update_start() then write the HTML file then update_end(). Never write the file without those flanking calls.
  • Configuration only through Settings in config.py (pydantic-settings, HTMLEDITOR_*). No os.environ read anywhere else.
  • Logging: logger = logging.getLogger(__name__), % lazy formatting, console on stderr. click.echo is for user output only. Never log on stdout: the MCP protocol and the CLI output live there.
  • Tracing: wrap every external call in trace_span("category.operation", {...}). Never put document content, prompts or credentials in an attribute.
  • data-doc-type on <html> drives the mode, data-type="slide" needs data-id and data-title, document headings are semantic <h1> to <h5>.
  • Templates are read only through the server; templates/bootstrap/slides-ei-empty.html is generated, edit the EI reference then make bootstrap-ei. For the medical charter it is the reverse: the bootstrap <style> is the source, edit it then make sync-medical-css to propagate the copy into the two reference decks.
  • A medical deck cites every third party image (non empty .med-source in the footer band) and anonymizes every patient image (no name, initials, MRN, care date, full date of birth; DICOM header and burned in pixels cleaned): see skill/types/medical.md. Its flex rows (.med-strip, .med-split, .med-cols, .med-compare, .med-steps, .med-stats, .med-timeline, .med-section-figs) need an explicit width class (w-25 to w-70) on every child, otherwise the PPTX export flattens the row into a vertical stack.
  • After editing anything under templates/, run make install or export HTMLEDITOR_TEMPLATES_DIR=$PWD/templates, otherwise the installed copy wins.
  • Every slide template (current and future) must ship fullscreen support (:fullscreen CSS, #btn-present button, enterPresentation/exitPresentation/ updateFullscreenScale JS): see .agent_docs/html-conventions.md § Fullscreen. A presentation file missing it (e.g. created before this convention) is a bug to patch, not a valid variant.
  • Architecture diagrams (data-type="arch-diagram") with 4+ nodes or a multi-row flow: never write data-x/data-y by hand, author the declarative topology (arch-row / arch-node / arch-col / arch-edge / arch-lane / arch-spacer) and run mcp-htmleditor arch-layout (or the layout_arch_diagram MCP tool) to compute positions. See skill/types/arch-diagram.md and src/mcp_htmleditor/arch_layout.py. The old manual data-x/data-y format stays valid for 2-3 node diagrams only.
  • After every arch-layout run, spawn a dedicated review sub-agent against mcp-htmleditor arch-checklist before considering the diagram done (protocol: skill/workflow-arch-qa.md). The checklist itself is user-editable at ~/.config/mcp-htmleditor/arch-checks/arch-diagram-checklist.md, never in code.
  • .mcp_state.json is written next to the edited file. It is gitignored, never commit it.
  • JS has no build step: node --check src/mcp_htmleditor/static/*.js.
  • src/mcp_htmleditor/version.py stays committed with "dev"; the build overwrites it from the git tag.

Read the full file on GitHub · 107 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 · 107 lines · 1,640 tokens per session scan A 6d0b96377c82

Subscribe to this mod's changes

mcp-htmleditor AGENTS.md is an instructions file published in the GitHub repository smorand/mcp-htmleditor (0 stars, last pushed 11d ago), licensed MIT. It adds 1,640 tokens to every session, about $0.0082 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 instructions, from other repositories

vscode buildNext.instructions.md

Working notes and architecture documentation for the new esbuild-based build system in build/next. Use when making changes to the new build pipeline (transpile/bundle commands, NLS plugin, source-map handling, resource copying, or self-hosting watch tasks).

microsoft/vscode · 6,785 tokens

spec-kit AGENTS.md

AGENTS.md instructions for github/spec-kit, covering agents.md, about spec kit and specify, quickstart — add a new integration in 5 steps, integration architecture and integrationmanifest — file tracking.

github/spec-kit · 7,104 tokens

codex AGENTS.md

AGENTS.md instructions for openai/codex, covering rust/codex-rs, the codex-core crate, code review rules, crate api surface and model visible context.

openai/codex · 5,182 tokens

langchain AGENTS.md

AGENTS.md instructions for langchain-ai/langchain, covering global development guidelines for the langchain monorepo, corridor security analysis, project architecture and context, monorepo structure and development tools & commands.

langchain-ai/langchain · 4,345 tokens

vscode oss-third-party-notices.instructions.md

Instructions for microsoft/vscode, covering vs code oss third-party-notices pipeline, architecture, pipeline flow in ci, applying the notice (cutover) and fallback chain (never fail the build).

microsoft/vscode · 5,001 tokens

next.js AGENTS.md

Instructions for vercel/next.js, covering next.js development guide, codebase structure, monorepo overview, core package: packages/next and other important packages.

vercel/next.js · 7,296 tokens