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 ymo6/illustrator-figure-mcp --skill audit-scientific-figuregit clone --depth 1 https://github.com/ymo6/illustrator-figure-mcpWrote 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/ymo6/illustrator-figure-mcp/audit-scientific-figure)<a href="https://agentmods.dev/skills/ymo6/illustrator-figure-mcp/audit-scientific-figure"><img src="https://agentmods.dev/badge/skills/ymo6/illustrator-figure-mcp/audit-scientific-figure/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/ymo6/illustrator-figure-mcp/audit-scientific-figure"><img src="https://agentmods.dev/badge/skills/ymo6/illustrator-figure-mcp/audit-scientific-figure.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 | $0.00072 | $0.01314 |
| Opus 5.5 | $0.00029 | $0.00526 |
| Sonnet 5.5 | $0.00014 | $0.00263 |
| Haiku 4.5 | $0.00007 | $0.00131 |
Grade A, and why
audit-scientific-figure 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 2mo 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 — 98 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Audit Scientific Figure
Act as the Reviewer. Read evidence and issue findings; do not draw while reviewing. A successful tool call is not evidence that the figure is correct.
Collect both evidence channels
Neither channel is sufficient alone.
- Geometry (deterministic). Run
audit_figure. It proves internal correctness: artboard containment, degenerate dimensions, duplicate names, overset text, raster declarations, connector routing and bindings, repeated-series alignment, and trace signatures. Passdeclared_object_countfrom the draw log to enable the bulk trace check. - Appearance (judged). Export with
export(format: png,scale: 2) and compare against the matching reference crop at readable resolution. Always re-export; never reuse a stale image.
audit_figure cannot see resemblance, colour intent, font choice, or scientific meaning. The render cannot see structure. A traced figure scores perfectly on appearance and is worthless.
Use get_document_structure or find_objects when a finding needs an object inventory.
Review taxonomy
Review every region, and then the whole figure, for:
- scientific semantics, exact readable text, topology, and arrow direction;
- editability coverage and meaningful object hierarchy;
- raster irreducibility and decomposition metadata;
- geometry, repeated alignment, equal spacing, margins, and whitespace;
- text fit, wrapping, font hierarchy, and colour consistency;
- clipping, unintended overlap, z-order, and object bounds;
- arrowhead clearance, connector routes through objects, label intersection, and route crossings;
- reference correspondence, or design consistency when there is no reference.
Read the audit report properly
passed: truemeans no hard geometric failure, nothing more. Never report it as "the figure is correct".findings_truncated: truefails the audit: unseen findings may remain.object_counts.untagged_paths > 0means geometry exists that this session did not build and cannot verify. Investigate before approving.object_counts.taggedat 0 with a populated document means nothing is verifiable at all.- Each finding carries
evidence,correction, andacceptance. Pass theacceptancecondition through to the Corrector unchanged — it is what the next audit checks.
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.
- 2mo ago First seen · 98 lines · 72 tokens per session scan A b0a510878b4b
audit-scientific-figure is a skill published in the GitHub repository ymo6/illustrator-figure-mcp (0 stars, last pushed 2mo ago), licensed MIT. It adds 72 tokens to every session and 1,314 once invoked, about $0.0003 per session on Opus 5.5. 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.
audit-scientific-figure compared with skills from other repositories
| Mod | Type | Stars | Last push | Licence | Scan | Tokens | Works with |
|---|---|---|---|---|---|---|---|
| audit-scientific-figureymo6/illustrator-figure-mcp | Skill | 0 | 2mo ago | MIT | A | 72 | Claude Code, Codex |
| scientific-schematicsK-Dense-AI/scientific-agent-skills | Skill | 56k | 6d ago | MIT | A | 68 | Claude Code |
| paper-poster-htmlwanshuiyin/Auto-claude-code-research-in-sleep | Skill | 19k | 2d ago | MIT | A | 107 | Claude Code |
| latex-postersforyourhealth111-pixel/Vibe-Skills | Skill | 3.8k | today | Apache-2.0 | A | 53 | Claude Code |
| figure-composercompanion-inc/feynman | Skill | 10k | today | MIT | A | 42 | Claude Code, Codex |
| paperFigWUBING2023/PaperSpine | Skill | 6.2k | 25d ago | MIT | A | 126 | Claude Code, Codex |
| schematicssynthetic-sciences/openscience | Skill | 4.4k | yesterday | Apache-2.0 | A | 134 | Claude Code |