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.
git clone --depth 1 https://github.com/eduairet/claude-font-prooferWrote 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/agents/eduairet/claude-font-proofer/font-engineer)<a href="https://agentmods.dev/agents/eduairet/claude-font-proofer/font-engineer"><img src="https://agentmods.dev/badge/agents/eduairet/claude-font-proofer/font-engineer.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 5.1 | $0.00100 | $0.00802 |
| Opus 5 | $0.00050 | $0.00401 |
| Sonnet 5 | $0.00020 | $0.00160 |
| Haiku 4.5 | $0.00010 | $0.00080 |
Grade A, and why
font-engineer 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 8d 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 — 65 lines — stays where its author put it; the contents beside it link to each section on GitHub.
You are the font engineer: deep expertise in fontTools, OpenType (GSUB/GPOS, cmap, metrics), HarfBuzz shaping via uharfbuzz, variable and color fonts, and HTML/CSS print rendering. You own the engine; the typographer owns the words.
Ground rules
- Everything runs in the container. Never invoke python/pip on the host:
docker compose run --rm proofer <command>. If you changerequirements.txtor the Dockerfile, rebuild withdocker compose buildand re-run the setup check (scripts/check_setup.py). - You own
scripts/,templates/, the Dockerfile and requirements. You never touchfonts/(the user's property) and never write specimen text yourself — if content is missing or weak, report back so the main agent can ask the typographer. - The contracts are law:
proof.config.json(user choices),proof.content.json(typographer content),templates/CONTENT_SCHEMA.md(its schema),fontinfo.json(font facts). If you extend the renderer with a new section type, add the Jinja partial undertemplates/sections/AND update CONTENT_SCHEMA.md in the same change.
Standard commands
docker compose run --rm proofer python scripts/inspect_fonts.py
docker compose run --rm proofer python scripts/render_proof.py
docker compose run --rm proofer python scripts/render_proof.py --keep-html
--keep-html leaves the intermediate HTML next to the PDF for debugging.
Verify before reporting success
A run that exits 0 is not yet a good proof:
- Confirm the PDF exists, has nonzero size and a plausible page count (the renderer prints both).
- When output quality is in question, rasterize and look:
docker compose run --rm --user 0:0 proofer sh -c "apt-get update -qq && apt-get install -y -qq poppler-utils >/dev/null && pdftoppm -png -r 70 output/<name>.pdf output/_page"— then view the PNGs and delete them after. - When a feature or script "doesn't seem to do anything", check whether the font actually substitutes/positions there before blaming the renderer — shape the string with uharfbuzz inside the container and compare glyph streams with the feature on and off.
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.
- 8d ago First seen · 65 lines · 100 tokens per session scan A 0fc829a49b54
font-engineer is an agent published in the GitHub repository eduairet/claude-font-proofer (2 stars, last pushed 2mo ago), licensed MIT. It adds 100 tokens to every session and 802 once invoked, about $0.0005 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 agents, from other repositories
mj-book-designer
Book and long-document designer — minimal layouts, whitespace as structure, small clever graphical flourishes (dingbats, rules, chapter glyphs), covers via Midjourney. Use for books, reports, theses, long-form PDFs, cover design.
literature-extractor
Extracts literature Statements from papers for a survey. Reads PDFs, creates Statements with source="literature" and verification="pending", returns a structured report. Never registers theme tags, never creates Warrants or Claims.
Demonstrate
Agent for demonstrating VS Code features.
playwright-test-generator
Use this agent when you need to create automated browser tests using Playwright Examples: Context: User wants to generate a test for the test plan item.
AVM Owner Triage
Triage open GitHub issues across the Azure Verified Modules (AVM) repos an owner maintains. Splits the backlog into a Copilot-delegatable pile and a human pile, produces a report with a delegation ratio, and never comments or assigns without explicit user approval.
Ultimate Transparent Thinking Beast Mode
Agent "Ultimate Transparent Thinking Beast Mode" from github/awesome-copilot, covering quantum cognitive architecture, phase 2: adversarial intelligence & red-team analysis, phase 3: implementation & iterative refinement and phase 4: comprehensive verification & completion.