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 skills add KerberosClaw/kc_ai_skills --skill md2pdfgit clone --depth 1 https://github.com/KerberosClaw/kc_ai_skillsWrote 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/skills/kerberosclaw/kc_ai_skills/md2pdf)<a href="https://agentmods.dev/skills/kerberosclaw/kc_ai_skills/md2pdf"><img src="https://agentmods.dev/badge/skills/kerberosclaw/kc_ai_skills/md2pdf/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/skills/kerberosclaw/kc_ai_skills/md2pdf"><img src="https://agentmods.dev/badge/skills/kerberosclaw/kc_ai_skills/md2pdf.svg" alt="Reviewed on agentmods" width="80" height="20"></a>- NVIDIA SkillSpector warn
SkillSpector: 2 findings, up to medium
These are SkillSpector’s own severities. On a checked sample its high-severity flags on skills were ~96% false positives — a documented command, a public API, a “never do X” rule — so we show them as a caution to read, not a verdict. Why →
- medium Excessive Agency · line 239 Skill enables autonomous high-impact decisions without human-in-the-loop verification. Critical operations (destructive commands, financial transactions, data deletion) should require explicit user confirmation.Fix: Add human-in-the-loop confirmation for destructive, irreversible, or high-impact operations. Never auto-execute commands that modify files, send data, or alter system state.
- medium Excessive Agency · line 277 Skill enables autonomous high-impact decisions without human-in-the-loop verification. Critical operations (destructive commands, financial transactions, data deletion) should require explicit user confirmation.Fix: Add human-in-the-loop confirmation for destructive, irreversible, or high-impact operations. Never auto-execute commands that modify files, send data, or alter system state.
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 5.1 | $0.00123 | $0.04230 |
| Opus 5 | $0.00062 | $0.02115 |
| Sonnet 5 | $0.00025 | $0.00846 |
| Haiku 4.5 | $0.00012 | $0.00423 |
Grade A, and why
md2pdf 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 10d 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 — 293 lines — stays where its author put it; the contents beside it link to each section on GitHub.
md2pdf
You are a Markdown-to-PDF production assistant. You convert exactly one Markdown file at a time into a clean, publication-ready A4 PDF while preserving the original source file.
Trigger
/md2pdf path/to/file.md
Prerequisites Check
Before anything else, verify these tools exist. If any is missing, stop and show install commands:
# Check all three
which pandoc && which mmdc && which weasyprint
Missing tool install commands:
- pandoc:
brew install pandoc - mmdc:
npm install -g @mermaid-js/mermaid-cli - weasyprint:
pip install weasyprintorbrew install weasyprint
Workflow
Step 1: Check for existing _pdf.md
If {filename}_pdf.md already exists, ask the user:
- Use existing: convert
_pdf.mddirectly to PDF (user may have manually tuned it) - Regenerate: copy from original and redo all conversions
Step 2: Ask style, and ask about a table of contents
2a. CSS style. Present options; if the user doesn't choose, pick the most suitable one automatically:
- Professional — dark blue headers, gray alternating rows, blue accent blockquotes (good for client-facing docs)
- Technical — compact, orange accent blockquotes, smaller fonts (good for dev manuals)
- Minimal — black and white, no colored headers (good for printing)
2b. Table of contents — always ask, never assume. A contents page also pulls the title onto a cover sheet, so switching it on costs a full page before the reader reaches any content. That is right for a report and absurd for a two-page memo.
Ask outright. If the user has no opinion, decide by length and say which you picked:
| Source length | Default | Why |
|---|---|---|
| Under ~6 pages of content | No TOC | A cover plus a contents list for a handful of sections is pure overhead |
Longer, or many ## sections |
TOC | Real page numbers make it navigable |
2c. Version footer. If the document carries a > Version: ... (or > 版本：...)
line under its H1, it is lifted out of the body and printed bottom-right on every
page. If the document is going to a third party and has no version line, offer to
add one — without it nobody can tell which draft they are holding.
What ships with it
3 files beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.
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.
- 10d ago First seen · 293 lines · 123 tokens per session scan A 46482401a1d9
md2pdf is a skill published in the GitHub repository KerberosClaw/kc_ai_skills (79 stars, last pushed 2d ago), licensed MIT. It adds 123 tokens to every session and 4,230 once invoked, about $0.0006 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 2026-08-30.
Other skills, from other repositories
tageszeitung
Creates a personalised daily newspaper from RSS feeds and web sources. Ported from the BACH news system (news.py + newspapergenerator.py). Own SQLite store (no Origin-DB). feedparser optional — XML fallback via stdlib. PDF export via Edge Headless (msedge.exe).
docutranslate
Use when translating documents locally via LLM — PDF, Word, Excel, Markdown, SRT subtitles with format preservation. DocuTranslate: LLM-powered multi-format local file translation tool with MCP server support.
llm-text-hygiene
Skill "llm-text-hygiene" from ellmos-ai/skills, covering übersicht & zweck, prüfkatalog, 2. platzhalter und baustellen-marker (eindeutig → auflösen), 4. ai-disclosure (prüfen → korrigieren) and ablauf.
steuer-assistent
Points to the standalone module steuer-assistent: a local, offline-first receipt worksheet for German employee income-related expenses (Werbungskosten) -- record, sum to the cent, private ZIP export. Use this skill when Werbungskosten receipts should be prepared in a structured way -- with a clear boundary: not tax…
finanz-versicherung
Strukturiert bereitgestellte Finanz- und Versicherungsunterlagen als neutrale Übersicht und Checkliste.
typst-cards
Generate PNG images and PDF carousels for online communication — social media, carousels, infographics, posts — using Typst. Use this skill whenever the user wants to create slides, cards, visual posts or any digital graphic content, even if they don't explicitly mention Typst. The skill drives an interview about…