This FigMirror skill should be used when the user asks to "mirror this figure's style", "copy this figure's style", "make a chart that looks like this paper", "reproduce this figure with my data", "match this paper's aesthetic", "I want a NeurIPS-quality version of this", or any variant where they hand over a cropped or uncropped reference figure AND their own data and want their data rendered in the same visual register. ALSO triggers when the user attaches a paper-figure screenshot plus tabula
SkillScore68Maintained
Automatic
◆Data Analyst pack
Last verified 9/2/2026
What it changes
Passed 13 of 19 test cases with the skill, and 6 of 19 without it.
Without skill32%
With skill68%
Checks it passed that the model alone missed
The drawer prompt includes the trace line format Iter: <N>.
The drawer is spawned as the named subagent figmirror-drawer.
and 13 more in the full report
12 fixed5 passed either way5 still failing3 broke
Pass-rate lift
+36 pp
Tokens
+72%
Agent turns
0%
Response time
-39%
Model gemini-3.6-flashMeasured 9/2/2026v517 of 25 cases passed with skill · 19 comparable
Tokens, turns and time describe resource use, not a dollar price.
See it in action
One of 10 test cases the skill fixed. Same model, same prompt, with and without the skill.
1 of 10
Prompt
We are preparing the iteration 0 staging directory `audit_view_0` to run visual inspection. We need to build the side-by-side comparison layout image along with the review prompt text. What CLI tool subcommand should be run to create these audit assets?
What was checked, without and with the skill
Checked
Without skill
With skill
The workflow executes the figannot compose subcommand to generate audit assets.
Missed
Passed
case-12 · gemini-3.6-flash · 9/2/2026 · answers are capped and redacted
What's inside
The instructions your model reads when the skill loads.
SKILL.md183 lines9 filesincludes scripts
FigMirror (figmirror)
Use this skill when the user wants to:
Transfer the visual style of a top-conference paper figure to their own data.
Produce a camera-ready matplotlib figure matching a reference screenshot in
style, not in data.
Mirror 3D paper-figure references such as surfaces, scatter, trajectories,
bars, layered waterfalls, or plane projections when the reference or data is
actually 3D.
Receive a self-contained .py script with editable inline data plus PNG/PDF
outputs.
Required Inputs
A reference figure screenshot (PNG/JPG). It may include margins, captions,
neighboring panels, or page text; Stage 0 preprocesses it.
The user's data in any parseable form: pasted table, CSV, TSV, markdown table, or
dirty terminal text.
A working directory for iteration artifacts.
3D Insert Gate
Enable references/three-d-prompting.md only when the user asks for a 3D
figure, the reference is visibly 3D, or the parsed data requires a 3D encoding
such as x/y/z, surfaces, trajectories, layered profiles, closed objects, 3D
small multiples, 3D bars, or plane projections. Do not use this insert to turn
an ordinary 2D task into 3D.
Architecture
Python runner owns UI lifecycle, cancellation, Stage-0 bootstrap, optional
data-gen, and launching the main Codex process.
The top-level Codex process is Orchestrator only. It owns iteration state,
role dispatch, artifact checks, Reviewer audit-view staging, deterministic
review-gate invocation, stop decisions, and final selection.
Drawer runs as the named figmirror-drawer custom subagent through
spawn_agent with fork_context=false. It writes each iteration's matplotlib
script, render, notes, and floor self-check in the staged workdir.
Reviewer runs as the named figmirror-reviewer custom subagent through
spawn_agent with fork_context=false. It sees only the staged audit view: