ワンクリックで
general-presentation
Generate and iteratively refine PowerPoint presentations from simulation results using python-pptx.
Codex または Claude でインストール この Prompt をコピーして Codex、Claude、または他のアシスタントに貼り付けると、Skill ページを確認してインストールできます。
メニュー
Generate and iteratively refine PowerPoint presentations from simulation results using python-pptx.
Codex または Claude でインストール この Prompt をコピーして Codex、Claude、または他のアシスタントに貼り付けると、Skill ページを確認してインストールできます。
SOC 職業分類に基づく
Quantify prediction uncertainty of MACE MLIPs using committee (ensemble) models; flag high-uncertainty structures for DFT verification.
Compute phonon-limited carrier mobility and mode-resolved electron-phonon coupling in 2D materials from first principles with Quantum ESPRESSO and EPW.
GPU-accelerated batched inference for MACE, MatGL (TensorNet/M3GNet/CHGNet), and FairChem MLIPs using NValchemi, enabling parallel static, relax, and MD workflows across multiple structures simultaneously.
Extract structured synthesis procedures from a folder of PDFs using the LeMat-Synth GeneralSynthesisOntology schema, producing one JSON file per paper with per-material synthesis records.
Predict LC-MS/MS (MS2, tandem mass spectra) from SMILES via ICEBERG, a two-stage deep neural network. Outputs predicted m/z vs intensity spectrum, fragment ion SMILES, and a spectrum plot.
Match an experimental spectrum (1H NMR, 13C NMR, IR) against predicted or database reference spectra for candidate ranking and structure confirmation. Supports local catalog lookup, public database fallback, and pluggable similarity metrics.
| name | general-presentation |
| description | Generate and iteratively refine PowerPoint presentations from simulation results using python-pptx. |
| category | ["general"] |
Create professional PowerPoint presentations from simulation results (plots, tables, parameters) using python-pptx. The agent writes standalone Python scripts using a helper library, enabling direct iteration with the user on the script to refine slides.
All scripts should import slide_utils from this skill:
# Env: base-agent
import sys
sys.path.insert(0, ".agents/skills/general-presentation/scripts")
from slide_utils import *
Available functions:
| Function | Purpose |
|---|---|
create_presentation(title, subtitle, author) | Create a new presentation with a styled title slide |
add_title_slide(prs, title, subtitle) | Additional title/divider slides |
add_section_slide(prs, title) | Section divider (colored background) |
add_image_slide(prs, title, image_path, caption, notes) | Single image/plot slide |
add_two_image_slide(prs, title, left_img, right_img, ...) | Side-by-side images |
add_image_and_text_slide(prs, title, image_path, text, ...) | Image + text layout |
add_table_slide(prs, title, headers, rows) | Data table slide |
add_bullets_slide(prs, title, bullets) | Bullet point slide |
add_autofit_box(slide, left, top, height, text, bg_color) | Add a colored box that auto-scales width to text |
save_presentation(prs, path) | Save to .pptx file |
Every builder function returns the Slide (or Shape) object, allowing further customization with raw python-pptx calls if needed.
Override the THEME dictionary before building slides:
from slide_utils import *
THEME["primary"] = RGBColor(0x00, 0x50, 0x80)
THEME["font_family"] = "Arial"
save_presentation(prs, "output.pptx")
The agent edits the Python script and re-runs to reflect changes. CRITICAL ITERATION LOOP: The agent MUST visually inspect the generated slides using their VLM capabilities to ensure elements do not overlap, text fits perfectly within boxes, and the slide space is used efficiently with a universal, professional design style.
To preview a slide as an image for visual inspection:
# Render PPTX to PDF using LibreOffice, then to PNG
libreoffice --headless --convert-to pdf output.pptx --outdir .
pdftocairo -png output.pdf output
# This generates output-1.png, which you can visually inspect.
Generating a simulation report:
# Env: base-agent
import sys
sys.path.insert(0, ".agents/skills/general-presentation/scripts")
from slide_utils import *
prs = create_presentation("LiFePO4 Stability", "MACE-MP Results", "Research Agent")
add_image_slide(prs, "Phonon DOS", "phonon_dos.png", caption="No imaginary modes observed")
add_table_slide(prs, "Parameters", ["Param", "Value"], [["Model", "MACE-MP"], ["fmax", "0.02 eV/A"]])
add_bullets_slide(prs, "Conclusions", ["Dynamically stable", "E_hull = 0 meV/atom"])
save_presentation(prs, "LiFePO4_report.pptx")
See amorphorization example for a complete runnable example with structure images, tables, and RDF plots.
base-agent with python-pptx installed.THEME["width"] / THEME["height"].Author: Bowen Deng Contact: GitHub @learningmatter-mit