Plotting Agent (Step 2)
Faithful implementation of the Plotting Agent from PaperOrchestra
(Song et al., 2026, arXiv:2604.05018, §4 Step 2 and App. F.1 p.45).
Cost: ~20–30 LLM calls. The paper uses PaperBanana (Zhu et al., 2026) as
the default backbone with a closed-loop VLM-critique refinement. This skill
expresses that loop in host-agent terms: you (the host agent) generate
matplotlib code with your own LLM, render via your Bash/Python tool,
optionally critique the rendered PNG with your vision model, redraw, and
finally caption.
Inputs
workspace/outline.json — specifically the plotting_plan array
workspace/inputs/idea.md and workspace/inputs/experimental_log.md —
the source data
workspace/inputs/figures/ — optional pre-existing figures (PlotOn mode)
Outputs
workspace/figures/<figure_id>.png — one PNG per plotting_plan entry
(300 DPI, sized to the requested aspect ratio)
workspace/figures/captions.json — {figure_id: caption_text} map
Workflow
Per figure (executed independently per figure_id)
Read the figure spec from outline.json:
{
"figure_id": "fig_main_results",
"title": "Main Results on Dataset X",
"plot_type": "plot",
"data_source": "experimental_log.md",
"objective": "Visual summary (Grouped Bar Chart) demonstrating ...",
"aspect_ratio": "5:4"
}
Few-shot retrieval (visual planning): pick the matching pattern from
references/chart-patterns.md (for plot_type=="plot") or
references/diagram-patterns.md (for plot_type=="diagram").
Extract data: parse idea.md and/or experimental_log.md
(data_source field tells you which) to obtain the numeric values or
conceptual entities the figure needs. For experimental_log.md, the
## 2. Raw Numeric Data section contains markdown tables.
Render:
If PAPERBANANA_PATH is set — use the PaperBanana backbone
(Zhu et al., 2026). It runs a Retriever → Planner → Stylist → Visualizer
→ Critic loop and is especially good for plot_type == "diagram".
See references/paperbanana-cookbook.md for setup (needs a Gemini API key).
python skills/plotting-agent/scripts/paperbanana_render.py \
--figure-id <figure_id> \
--caption "<objective from figure spec>" \
--content-file workspace/inputs/idea.md \
--task <diagram|plot> \
--aspect-ratio <aspect_ratio> \
--out workspace/figures/<figure_id>.png
Otherwise — write a matplotlib script and run it via your Bash tool,
or use the bundled helper:
python skills/plotting-agent/scripts/render_matplotlib.py \
--spec spec.json \
--out workspace/figures/<figure_id>.png
The script must apply the academic style from chart-patterns.md, use the
correct pixel size from aspect-ratios.md, save at 300 DPI, and call
plt.close() after savefig.
VLM critique loop (optional, only if your host has vision):
- Reload the rendered PNG as a multimodal input to your LLM.
- Critique it against the figure's
objective from the outline. Look for:
visual artifacts, mislabeled axes, illegible text, color clashes,
misleading scaling, missing legend, overlapping labels.
- If problems are found, regenerate the matplotlib script with corrections
and re-render. Cap at 3 critique iterations per figure.
- This is the closed-loop refinement step the paper inherits from
PaperBanana. See
references/plotting-pipeline.md for the full loop
description.
- If your host has no vision input, skip this step entirely. The
figure will still render correctly, just without iterative refinement.
Generate the caption using the verbatim Caption Generation prompt at
references/caption-prompt.md. Inputs to the caption prompt:
task_name — the section the figure belongs to (e.g., "Methodology",
"Experiments")
raw_content — the surrounding section text (or content_bullets from
the section_plan if the section isn't drafted yet)
description — the objective field from the figure spec
figure_desc — a 1-sentence description of what the rendered figure
actually shows (from your VLM critique pass, or from the script's plan
if no vision)
Write the caption to workspace/figures/captions.json keyed by
figure_id. Captions must NOT contain Figure N: or Caption N:
prefixes — the LaTeX template handles numbering. Plain text only, no
markdown.
Conceptual diagrams
For plot_type == "diagram", prefer PaperBanana when available — its
Retriever grounds the Planner in real published paper diagrams. If
PAPERBANANA_PATH is unset, follow references/diagram-patterns.md.
Patterns include block diagrams, system overviews, flowcharts, and
algorithm-as-graph. The bundled helper:
python skills/plotting-agent/scripts/render_diagram.py \
--spec diagram_spec.json \
--out workspace/figures/<figure_id>.png
handles the simple cases (boxes-and-arrows). For complex Fig-1-style
overview diagrams, write matplotlib patches code yourself.
Hard rules
- 300 DPI for every figure. Lower DPI gets rejected at the LaTeX compile
step on conference templates.
- Aspect ratio is exact. The figure spec's
aspect_ratio is one of 12
enumerated strings. Use the pixel targets in references/aspect-ratios.md.
- Hide top and right spines for plots. (Diagrams: no spines at all.)
- Muted academic colors only. The palette is in
chart-patterns.md.
Never use matplotlib defaults (too saturated for print).
- No 3D, no pie charts, no decorative visuals. The paper's evaluators
penalize these.
- Every figure MUST have a caption in
captions.json. The Section
Writing Agent will fail-stop if a caption is missing for any figure
referenced from the outline.
- No
Figure N: prefix in captions — LaTeX adds it.
- Never describe data you didn't plot. The Plotting Agent must not
hallucinate axes, baselines, or trends. Source-of-truth is
experimental_log.md or idea.md.
Pre-existing figures (PlotOn mode)
If workspace/inputs/figures/ is non-empty, check whether any pre-existing
file matches a figure_id in the outline (by filename prefix). If so,
copy it into workspace/figures/ as-is and still generate a caption
using the caption prompt. Only generate from scratch the figure_ids that
have no pre-existing counterpart.
Resources
references/caption-prompt.md — verbatim Caption Generation prompt from App. F.1
references/plotting-pipeline.md — the full few-shot → render → critique → caption loop
references/chart-patterns.md — matplotlib style + chart type recipes
references/diagram-patterns.md — conceptual diagram recipes
references/aspect-ratios.md — pixel targets for each of the 12 allowed ratios at 300 DPI
references/paperbanana-cookbook.md — NEW PaperBanana setup, usage, cost notes, attribution
scripts/render_matplotlib.py — render a JSON plot spec → PNG (matplotlib fallback)
scripts/render_diagram.py — render a JSON diagram spec → PNG (matplotlib fallback)
scripts/paperbanana_render.py — NEW PaperBanana backbone wrapper (reads PAPERBANANA_PATH from env)
1---2name: plotting-agent3description: Step 2 of the PaperOrchestra pipeline (arXiv:2604.05018). Execute the visualization plan from outline.json — render plots and conceptual diagrams from experimental_log.md and idea.md, optionally refine via VLM critique loop, and produce context-aware captions. Runs in parallel with the literature-review-agent. TRIGGER when the orchestrator delegates Step 2 or when the user asks to "generate the figures for my paper" or "render the plots from this experiment log".4---56# Plotting Agent (Step 2)78Faithful implementation of the Plotting Agent from PaperOrchestra9(Song et al., 2026, arXiv:2604.05018, §4 Step 2 and App. F.1 p.45).1011**Cost: ~20–30 LLM calls.** The paper uses PaperBanana (Zhu et al., 2026) as12the default backbone with a closed-loop VLM-critique refinement. This skill13expresses that loop in host-agent terms: you (the host agent) generate14matplotlib code with your own LLM, render via your Bash/Python tool,15optionally critique the rendered PNG with your vision model, redraw, and16finally caption.1718## Inputs1920- `workspace/outline.json` — specifically the `plotting_plan` array21- `workspace/inputs/idea.md` and `workspace/inputs/experimental_log.md` —22 the source data23- `workspace/inputs/figures/` — optional pre-existing figures (`PlotOn` mode)2425## Outputs2627- `workspace/figures/<figure_id>.png` — one PNG per `plotting_plan` entry28 (300 DPI, sized to the requested aspect ratio)29- `workspace/figures/captions.json` — `{figure_id: caption_text}` map3031## Workflow3233### Per figure (executed independently per `figure_id`)34351. **Read the figure spec** from `outline.json`:36 ```json37 {38 "figure_id": "fig_main_results",39 "title": "Main Results on Dataset X",40 "plot_type": "plot",41 "data_source": "experimental_log.md",42 "objective": "Visual summary (Grouped Bar Chart) demonstrating ...",43 "aspect_ratio": "5:4"44 }45 ```46472. **Few-shot retrieval (visual planning)**: pick the matching pattern from48 `references/chart-patterns.md` (for `plot_type=="plot"`) or49 `references/diagram-patterns.md` (for `plot_type=="diagram"`).50513. **Extract data**: parse `idea.md` and/or `experimental_log.md`52 (`data_source` field tells you which) to obtain the numeric values or53 conceptual entities the figure needs. For `experimental_log.md`, the54 `## 2. Raw Numeric Data` section contains markdown tables.55564. **Render**:5758 **If `PAPERBANANA_PATH` is set** — use the PaperBanana backbone59 (Zhu et al., 2026). It runs a Retriever → Planner → Stylist → Visualizer60 → Critic loop and is especially good for `plot_type == "diagram"`.61 See `references/paperbanana-cookbook.md` for setup (needs a Gemini API key).6263 ```bash64 python skills/plotting-agent/scripts/paperbanana_render.py \65 --figure-id <figure_id> \66 --caption "<objective from figure spec>" \67 --content-file workspace/inputs/idea.md \68 --task <diagram|plot> \69 --aspect-ratio <aspect_ratio> \70 --out workspace/figures/<figure_id>.png71 ```7273 **Otherwise** — write a matplotlib script and run it via your Bash tool,74 or use the bundled helper:75 ```bash76 python skills/plotting-agent/scripts/render_matplotlib.py \77 --spec spec.json \78 --out workspace/figures/<figure_id>.png79 ```80 The script must apply the academic style from `chart-patterns.md`, use the81 correct pixel size from `aspect-ratios.md`, save at 300 DPI, and call82 `plt.close()` after `savefig`.83845. **VLM critique loop (optional, only if your host has vision)**:85 - Reload the rendered PNG as a multimodal input to your LLM.86 - Critique it against the figure's `objective` from the outline. Look for:87 visual artifacts, mislabeled axes, illegible text, color clashes,88 misleading scaling, missing legend, overlapping labels.89 - If problems are found, regenerate the matplotlib script with corrections90 and re-render. Cap at 3 critique iterations per figure.91 - This is the closed-loop refinement step the paper inherits from92 PaperBanana. See `references/plotting-pipeline.md` for the full loop93 description.94 - **If your host has no vision input, skip this step entirely.** The95 figure will still render correctly, just without iterative refinement.96976. **Generate the caption** using the verbatim Caption Generation prompt at98 `references/caption-prompt.md`. Inputs to the caption prompt:99 - `task_name` — the section the figure belongs to (e.g., "Methodology",100 "Experiments")101 - `raw_content` — the surrounding section text (or content_bullets from102 the section_plan if the section isn't drafted yet)103 - `description` — the `objective` field from the figure spec104 - `figure_desc` — a 1-sentence description of what the rendered figure105 actually shows (from your VLM critique pass, or from the script's plan106 if no vision)107108 Write the caption to `workspace/figures/captions.json` keyed by109 `figure_id`. **Captions must NOT contain `Figure N:` or `Caption N:`110 prefixes** — the LaTeX template handles numbering. Plain text only, no111 markdown.112113## Conceptual diagrams114115For `plot_type == "diagram"`, prefer PaperBanana when available — its116Retriever grounds the Planner in real published paper diagrams. If117`PAPERBANANA_PATH` is unset, follow `references/diagram-patterns.md`.118Patterns include block diagrams, system overviews, flowcharts, and119algorithm-as-graph. The bundled helper:120121```bash122python skills/plotting-agent/scripts/render_diagram.py \123 --spec diagram_spec.json \124 --out workspace/figures/<figure_id>.png125```126127handles the simple cases (boxes-and-arrows). For complex Fig-1-style128overview diagrams, write matplotlib patches code yourself.129130## Hard rules131132- **300 DPI** for every figure. Lower DPI gets rejected at the LaTeX compile133 step on conference templates.134- **Aspect ratio is exact**. The figure spec's `aspect_ratio` is one of 12135 enumerated strings. Use the pixel targets in `references/aspect-ratios.md`.136- **Hide top and right spines** for plots. (Diagrams: no spines at all.)137- **Muted academic colors** only. The palette is in `chart-patterns.md`.138 Never use matplotlib defaults (too saturated for print).139- **No 3D, no pie charts, no decorative visuals.** The paper's evaluators140 penalize these.141- **Every figure MUST have a caption** in `captions.json`. The Section142 Writing Agent will fail-stop if a caption is missing for any figure143 referenced from the outline.144- **No `Figure N:` prefix** in captions — LaTeX adds it.145- **Never describe data you didn't plot.** The Plotting Agent must not146 hallucinate axes, baselines, or trends. Source-of-truth is147 `experimental_log.md` or `idea.md`.148149## Pre-existing figures (PlotOn mode)150151If `workspace/inputs/figures/` is non-empty, check whether any pre-existing152file matches a `figure_id` in the outline (by filename prefix). If so,153**copy** it into `workspace/figures/` as-is and **still generate a caption**154using the caption prompt. Only generate from scratch the figure_ids that155have no pre-existing counterpart.156157## Resources158159- `references/caption-prompt.md` — verbatim Caption Generation prompt from App. F.1160- `references/plotting-pipeline.md` — the full few-shot → render → critique → caption loop161- `references/chart-patterns.md` — matplotlib style + chart type recipes162- `references/diagram-patterns.md` — conceptual diagram recipes163- `references/aspect-ratios.md` — pixel targets for each of the 12 allowed ratios at 300 DPI164- `references/paperbanana-cookbook.md` — **NEW** PaperBanana setup, usage, cost notes, attribution165- `scripts/render_matplotlib.py` — render a JSON plot spec → PNG (matplotlib fallback)166- `scripts/render_diagram.py` — render a JSON diagram spec → PNG (matplotlib fallback)167- `scripts/paperbanana_render.py` — **NEW** PaperBanana backbone wrapper (reads `PAPERBANANA_PATH` from env)