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 commands/mfmezger/ai_agent_dotfiles/html-reportinggit clone --depth 1 https://github.com/mfmezger/ai_agent_dotfilesWrote 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/commands/mfmezger/ai_agent_dotfiles/html-reporting)<a href="https://agentmods.dev/commands/mfmezger/ai_agent_dotfiles/html-reporting"><img src="https://agentmods.dev/badge/commands/mfmezger/ai_agent_dotfiles/html-reporting.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 | $0.00185 | $0.04018 |
| Opus 5 | $0.00093 | $0.02009 |
| Sonnet 5 | $0.00037 | $0.00804 |
| Haiku 4.5 | $0.00018 | $0.00402 |
Grade A, and why
html-reporting 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 — 229 lines — stays where its author put it; the contents beside it link to each section on GitHub.
HTML Reporting
Markdown is the default agent output format, but it caps out fast. Long Markdown rarely gets read; HTML does. This skill is about producing single-file HTML artifacts — reports, specs, PR explainers, prototypes, custom editors — that humans actually open, scroll through, and share.
Why HTML beats Markdown for human-facing output
Markdown's killer feature was always "easy to hand-edit." That feature has lost most of its value: people now edit by prompting the agent to change the file rather than typing into it themselves. What remains is reading and sharing, and Markdown is bad at both once the document gets long.
HTML wins on the dimensions that actually matter for finished artifacts:
- Information density. Tables, SVG diagrams, color coding, code with syntax highlighting, side-by-side comparisons, tabs. A 200-line Markdown plan becomes a navigable HTML page with a TOC and collapsible sections.
- Readability at length. Past ~100 lines, Markdown becomes a wall. Real typography, spacing, and visual hierarchy make HTML scannable.
- Sharing. Drop the file into S3, GitHub Pages, or any static host and send a URL. Browsers render it everywhere, including on phones. Markdown attachments get ignored.
- Interactivity. Sliders, knobs, drag-and-drop, live previews, "copy as JSON" buttons — all trivially possible, and they close the loop back to the agent.
- Diagrams. When HTML is on the table, the model draws SVG diagrams instead of butchering ASCII art or Unicode color blocks.
The tradeoff: HTML uses roughly 2–4× more tokens and takes proportionally longer to generate. For anything a human will read more than once, that cost is trivially worth it.
When to reach for this skill
Use HTML when any of these are true:
- A human will read this more than they will edit it (specs, reports, plans, reviews, research).
- The output would exceed ~100 lines as Markdown.
- It needs visual structure: tables, diagrams, color coding, comparisons, annotations.
- It needs interactivity: tunable parameters, drag-and-drop, live previews, export buttons.
- It will be shared with teammates or leadership.
- The Markdown version would force ASCII-art or Unicode "color block" hacks.
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 · 229 lines · 185 tokens per session scan A 1ff61c136d65
html-reporting is a command published in the GitHub repository mfmezger/ai_agent_dotfiles (5 stars, last pushed 1mo ago), licensed MIT. It adds 185 tokens to every session and 4,018 once invoked, about $0.0009 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 commands, from other repositories
annotate
The /plannotator-annotate slash command for annotating markdown files, HTML files, URLs, and folders.
docs-builder
Reorg a docs corpus, split an oversized doc, search it, keep pages current, index them.
wtfp:export-latex
Export the manuscript and bibliography to portable LaTeX source with declared formatting.
brand-generate
Generate an on-brand document from a saved Brand Profile.
ppt-image2-editable-rebuild
Rebuild image2 or imagegen reference slides as editable PowerPoint decks.
convert
Convert a file to Markdown using markitdown.