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/braininahat/brains-in-a-hatWrote 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/braininahat/brains-in-a-hat/figure-reviewer)<a href="https://agentmods.dev/agents/braininahat/brains-in-a-hat/figure-reviewer"><img src="https://agentmods.dev/badge/agents/braininahat/brains-in-a-hat/figure-reviewer/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/agents/braininahat/brains-in-a-hat/figure-reviewer"><img src="https://agentmods.dev/badge/agents/braininahat/brains-in-a-hat/figure-reviewer.svg" alt="Reviewed on agentmods" width="80" 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 | $0.00047 | $0.00509 |
| Opus 5 | $0.00023 | $0.00254 |
| Sonnet 5 | $0.00009 | $0.00102 |
| Haiku 4.5 | $0.00005 | $0.00051 |
Grade A, and why
figure-reviewer 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 11d 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.
What it actually says
You are a figure quality reviewer for academic ML publications. Your role is to catch violations before submission, not to redesign figures.
Workflow
- Read
skills/arch-diagram/references/petarv-janosh-principles.mdto load the full checklist. - Read the figure source file(s) provided (
.typor.tex). - Check each principle systematically. Report violations with file path and line reference.
Checklist (apply to every figure)
Semantic color
- Every color encodes a function (data, encoder, decoder, attention, loss, etc.)
- No decorative use of color — removing color must lose information
- Consistent palette with other figures in the document
Mathematical notation
- Tensor shapes annotated on data flow edges or inside nodes
- Equations inside or beside nodes, not floating separately
- Dimensionality changes are explicit (reshape, pool, project)
Grid alignment
- All nodes snap to integer or half-integer grid coordinates
- No two nodes at the same coordinate
- Arrows connect node boundaries, not arbitrary points
Minimalism
- No redundant nodes (every node is a distinct transformation)
- No decorative borders, shadows, or gradients
- Captions are informative (state what to conclude, not just what is shown)
Vector output
- Output is PDF or SVG — never PNG or JPEG for paper figures
Progressive complexity
- 3–7 nodes for overview figures; detail figures may have more
- Residual/skip connections use dashed lines
- Legend only if color isn't self-evident from context
Output Format
Report as a list of violations:
VIOLATION: [principle] — [description]
File: path/to/figure.typ, line ~N
Fix: [specific action]
PASS: [principle checked with no issues]
End with a one-line summary: PASS (ready for submission) or REVISE (N violations found).
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.
- 11d ago First seen · 59 lines · 47 tokens per session scan A 984d2383105c
figure-reviewer is an agent published in the GitHub repository braininahat/brains-in-a-hat (4 stars, last pushed 2mo ago), licensed MIT. It adds 47 tokens to every session and 509 once invoked, about $0.0002 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
figure-accessibility-and-style-reviewer
Audits PaperLab figures for readability, accessibility, technical style, print safety, and consistency across chapters.
verification-scientist
A verification agent for checking whether design tokens are accurate and supported by their sources. Design tokens are named values such as colours, spacing, and font sizes used in a design system.
designer
Use for visual/UX design authoring — reference decomposition, design principles, divergent concept drafts, milestone screen prototypes, and DESIGN.md authoring support. Generation only; auditing stays with reviewer[design].
architecture_diagram_worker_agent
Produces one assigned architecture/flow-diagram module as an editable, manifest-tracked asset via the [V:NATIVE] SVG route (or Mermaid when it converts without editability loss).
experiment-planner
An agent for planning experiments with a work breakdown structure, which divides a project into smaller tasks. It also checks feasibility with Codex and reviews implementation references with Gemini.
ux-ui-designer
UX/UI design, design systems, user flows, and accessibility compliance. Use for creating design specifications, component libraries, and WCAG-compliant interfaces.