Skip to content
Back to search

engineering-figure-agent

Engineering Figure Banana is a Codex skill for generating publication-style engineering and computer science figures, with support for Chinese academic papers, bilingual labeling, and editable-figure handoff workflows.

Safety scan

A static text scan, not a guarantee. Select the badge to see what was scanned and every finding.

Checked today
Badge for your README
findsafeskills safety rating

Markdown

[![findsafeskills safety rating](/api/listings/heyu-233--engineering-figure-agent/badge.svg)](/listings/heyu-233--engineering-figure-agent)

HTML

<a href="/listings/heyu-233--engineering-figure-agent"><img src="/api/listings/heyu-233--engineering-figure-agent/badge.svg" alt="findsafeskills safety rating" /></a>

Install

i

Install instructions from the README

As written by the repo's author. findsafeskills has not verified these steps.

Install and run the setup check:

& "$HOME/.codex/skills/engineering-figure-agent/scripts/install_and_test.ps1" -RunSetupCheck

Open the wizard:

& "$HOME/.codex/skills/engineering-figure-agent/scripts/wizard.ps1"

Or generate a conceptual figure directly:

python "$HOME/.codex/skills/engineering-figure-agent/scripts/generate_image.py" `
  --figure-template system-architecture `
  --lang en `
  "A retrieval-augmented generation system with OCR, chunking, embedding, vector search, reranking, and answer synthesis."

For installation details, use the language-specific guides:

- [README.zh-CN.md](./README.zh-CN.md): Windows 最短安装路径、验证方法、中文示例 - [README.en.md](./README.en.md): setup, verification, examples, positioning

What was scanned

The files the safety scan read, as written by the repository's author and shown as plain text. Lines with a finding are marked.

SKILL.md · Manifest · 97 lines

View this file on GitHub (read at the repository's current version: the scan did not record which commit it read)

  1. ---
  2. name: engineering-figure-agent
  3. description: Use when the user needs engineering or research-paper figures: system architecture diagrams, algorithm workflows, hardware schematics, benchmark charts, ablation plots, figure briefs, redraws, image edits, or exact publication-style plots from numeric data. Covers Gemini, Nano Banana, OpenAI image models, and local plotting workflows.
  4. ---
  5. # Engineering Figure Agent
  6. Use this skill after the figure goal is already reasonably clear.
  7. ## Boundary
  8. Good fit:
  9. - Turn a figure brief into a conceptual diagram, workflow figure, schematic, or exact publication plot.
  10. - Choose between `image`, `plot`, and `mixed` mode.
  11. - Build prompts, render plots, and refine existing briefs.
  12. Not the main tool for:
  13. - Deciding from scratch what claim the paper should visualize.
  14. - Full paper-argument or reviewer-style critique.
  15. If the user is still deciding the figure claim, panel logic, or caption argument, use an available research-writing or paper-analysis skill upstream first.
  16. ## Core Decision
  17. - `image` mode: conceptual figures, architecture diagrams, algorithm workflows, graphical abstracts, schematics, reference-inspired redraws.
  18. - `plot` mode: exact bar charts, trend curves, heatmaps, scatter plots, ablation plots, benchmark summaries.
  19. - `mixed` mode: render quantitative panels locally first, then use image generation only for conceptual panels.
  20. Never use image generation for exact values, axes, or benchmark geometry.
  21. ## Default Workflow
  22. 1. Inspect the user input and decide whether a figure brief is already present.
  23. 2. If needed, create a brief using `references/figure-brief-spec.md`.
  24. 3. Choose `image`, `plot`, or `mixed` mode.
  25. 4. For conceptual figures, prefer the prompt-builder scripts; read template references only when you need to inspect or customize template wording.
  26. 5. For exact plots, create a concise plot request and render it locally.
  27. 6. Keep labels short, claims source-grounded, and outputs publication-oriented.
  28. 7. Save prompt/spec/output paths when files are produced.
  29. 8. Before finishing, run only the checks needed for the touched path.
  30. ## Reference Loading
  31. Read only what is needed:
  32. - `references/figure-brief-spec.md`: brief structure and mode rules.
  33. - `references/publication-figure-design.md`: publication styling defaults.
  34. - `references/provider-selection.md`: Gemini/Banana/OpenAI setup.
  35. - `references/highres-policy.md`: high-resolution and fail-closed rules.
  36. - `references/chinese-labels.md`: Chinese label readability rules.
  37. - `references/engineering-figure-templates.md` and `references/materials-science-figure-template.md`: template wording when script output needs customization.
  38. - `references/publication-chart-patterns.md`, `references/natural-language-plot-workflow.md`, and `references/publication-plot-api.md`: exact or plot-heavy workflows.
  39. - `references/editable-figure-handoff.md`: optional editable SVG handoff.
  40. ## Scripts
  41. - Build a prompt without network calls:
  42. ```bash
  43. python scripts/build_engineering_figure_prompt.py --figure-template system-architecture --lang en "technical background"
  44. python scripts/build_materials_figure_prompt.py --materials-figure mechanism-figure --lang en "scientific background"
  45. ```
  46. - Generate or edit a conceptual image:
  47. ```bash
  48. python scripts/generate_image.py --provider openai --figure-template system-architecture --lang en "technical background"
  49. python scripts/generate_image.py --provider openai --materials-figure mechanism-figure --lang en "scientific background"
  50. ```
  51. - Render an exact plot:
  52. ```bash
  53. python scripts/build_plot_spec.py request.json --out spec.json
  54. python scripts/plot_publication_figure.py spec.json --out-path output/figure --formats png pdf svg
  55. ```
  56. - Use the unified CLI:
  57. ```bash
  58. python scripts/efa.py prompt --figure-template system-architecture "technical background"
  59. python scripts/efa.py prompt --materials-figure mechanism-figure "scientific background"
  60. python scripts/efa.py plot request.json --out-path output/figure
  61. python scripts/efa.py check
  62. ```
  63. ## Quality Rules
  64. - Do not fabricate measurements, benchmark values, hardware specs, or unsupported causal claims.
  65. - Keep arrows, reading order, and module hierarchy explicit.
  66. - Favor white backgrounds, publication-style spacing, and readable labels.
  67. - Use prompt-first workflows when layout fidelity matters.
  68. - For high-resolution requests, follow `references/highres-policy.md`.
  69. - For third-party providers, require explicit opt-in before sending keys or user files.
  70. - For Chinese figures, preserve standard technical notation when it improves clarity.
README.md · README · 192 lines

View this file on GitHub (read at the repository's current version: the scan did not record which commit it read)

  1. # Engineering Figure Agent
  2. <div align="center">
  3. ![License](https://img.shields.io/badge/license-MIT-2563eb)
  4. ![Codex Skill](https://img.shields.io/badge/Codex-skill-111827)
  5. ![Figure Modes](https://img.shields.io/badge/modes-image%20%7C%20plot-f59e0b)
  6. ![Backends](https://img.shields.io/badge/backends-Gemini%20%7C%20OpenAI%20%7C%20local%20plots-16a34a)
  7. ![Focus](https://img.shields.io/badge/focus-engineering%20papers-7c3aed)
  8. [![Try Engineering Figure Agent on Socialistic](https://socialistic.ai/api/embed/engineering-figure-agent-407b80)](https://socialistic.ai/zh/skill/engineering-figure-agent-407b80?utm_source=github&utm_medium=readme&utm_campaign=20260519-student-research-tools&utm_content=badge)
  9. **Agent-native figure production for engineering and CS papers.**
  10. 把工程论文里的系统架构图、算法流程图、实验曲线和多面板图,拆成可控的生成流程:概念图走 image mode,精确数值图走 plot mode。
  11. [中文说明](./README.zh-CN.md) | [English Guide](./README.en.md) | [Example Gallery](./docs/examples/README.md) | [Showcase](./docs/showcase.md)
  12. </div>
  13. ## Preview
  14. | System / Architecture | Cooperative Perception | Safety Taxonomy |
  15. |---|---|---|
  16. | ![Federated open-vocabulary driving figure](docs/examples/federated-open-vocab-driving-2k-1.png) | ![Cooperative object tracking figure](docs/examples/cooperative-object-tracking-2k-1.png) | ![Multi-agent safety overview figure](docs/examples/multi-agent-safety-overview-2k-1.png) |
  17. | Dense Systems Diagram | Deployment Scenario |
  18. |---|---|
  19. | ![Linux kernel system diagram](docs/examples/linux-kernel-system-1.jpg) | ![Health monitoring and early warning deployment scenarios](docs/examples/health-monitoring-early-warning-reference.jpg) |
  20. | Exact Local Plot |
  21. |---|
  22. | ![Benchmark plot](docs/examples/benchmark-plot.png) |
  23. ## Why It Exists
  24. Most figure tools treat every paper figure as the same image prompt problem. Engineering papers do not work that way.
  25. | Figure need | Better path |
  26. |---|---|
  27. | System architecture, pipeline, schematic, graphical abstract | `image mode` with an engineering-aware prompt template |
  28. | Benchmark curves, ablation bars, heatmaps, scatter plots | `plot mode` with local plotting and exact values |
  29. | Mixed conceptual + quantitative figure | Render numeric panels locally, then compose or describe the conceptual panels |
  30. | Reference-inspired redraw | Use image mode for structure and style exploration, then manually verify labels and layout |
  31. Engineering Figure Agent is intentionally lighter than a full paper-upload platform. It is built for researchers who already know what claim a figure should support and want a cleaner production path inside an agent workflow.
  32. ## Quick Start
  33. Install and run the setup check:
  34. ```powershell
  35. & "$HOME/.codex/skills/engineering-figure-agent/scripts/install_and_test.ps1" -RunSetupCheck
  36. ```
  37. Open the wizard:
  38. ```powershell
  39. & "$HOME/.codex/skills/engineering-figure-agent/scripts/wizard.ps1"
  40. ```
  41. Or generate a conceptual figure directly:
  42. ```powershell
  43. python "$HOME/.codex/skills/engineering-figure-agent/scripts/generate_image.py" `
  44. --figure-template system-architecture `
  45. --lang en `
  46. "A retrieval-augmented generation system with OCR, chunking, embedding, vector search, reranking, and answer synthesis."
  47. ```
  48. For installation details, use the language-specific guides:
  49. - [README.zh-CN.md](./README.zh-CN.md): Windows 最短安装路径、验证方法、中文示例
  50. - [README.en.md](./README.en.md): setup, verification, examples, positioning
  51. ## Two Modes
  52. ### Image Mode
  53. Use `image mode` when visual structure matters more than exact numeric geometry.
  54. Best for:
  55. - system architecture diagrams
  56. - algorithm workflows
  57. - graphical abstracts
  58. - electronics or embedded-system schematics
  59. - reference-inspired redraws and layout exploration
  60. Supported conceptual-image backends:
  61. | Provider | Use case | Config |
  62. |---|---|---|
  63. | `gemini` / `banana` | Google Gemini / Banana-compatible generation | `NANOBANANA_*` variables |
  64. | `openai` | OpenAI Image API backend for conceptual figures and edits | `OPENAI_API_KEY` or `OPENAI_API_KEY_FILE` |
  65. Example:
  66. ```powershell
  67. python "$HOME/.codex/skills/engineering-figure-agent/scripts/generate_image.py" `
  68. --provider openai `
  69. --model gpt-image-1.5 `
  70. --figure-template system-architecture `
  71. --lang en `
  72. --openai-quality auto `
  73. --openai-size auto `
  74. "A retrieval-augmented generation system with OCR, chunking, embedding, vector search, reranking, and answer synthesis."
  75. ```
  76. ### Plot Mode
  77. Use `plot mode` when values, axes, and geometry must stay exact.
  78. Best for:
  79. - benchmark bar charts
  80. - ablation plots
  81. - trend curves
  82. - heatmaps
  83. - scatter plots
  84. - multi-panel quantitative figures
  85. Rule of thumb:
  86. | If the figure needs... | Use |
  87. |---|---|
  88. | exact values, axes, error bars, or benchmark geometry | `plot mode` |
  89. | concept explanation, layout exploration, or architecture visuals | `image mode` |
  90. | both | local plot panels first, image generation only for conceptual parts |
  91. ## Workflow
  92. Recommended upstream handoff:
  93. 1. Use `ai-research-writing-guide` to decide what claim the figure should support.
  94. 2. Write or collect the figure brief: audience, claim, modules, labels, data, and target style.
  95. 3. Use `engineering-figure-agent` to render the diagram or exact plot.
  96. 4. Verify labels, numeric truth, publication style, and export format before paper submission.
  97. ## Platform Adapters
  98. | Platform | Entry | Best use |
  99. |---|---|---|
  100. | Codex | [SKILL.md](./SKILL.md) | Full agent-native workflow with local scripts |
  101. | Claude Code | [adapters/claude-code/](./adapters/claude-code/) | Local repository figure work and prompt/plot generation |
  102. | ChatGPT / Claude web | [docs/prompt-pack.md](./docs/prompt-pack.md) | Chat-only figure brief and prompt drafting |
  103. | VS Code / Obsidian | [templates/figure-brief/](./templates/figure-brief/) | Edit and archive figure briefs, prompts, and plot requests |
  104. Core platform-neutral contracts:
  105. - [Figure brief spec](./references/figure-brief-spec.md)
  106. - [Plot request schema](./schemas/plot-request.schema.json)
  107. - [Figure brief schema](./schemas/figure-brief.schema.json)
  108. ## Support Matrix
  109. | Platform | Status | Notes |
  110. |---|---|---|
  111. | Windows | tested | primary tested platform; helper scripts supported first |
  112. | macOS | reported working | successful installs have been reported, including AI-assisted setup |
  113. | Linux | expected for core Python workflow | some environments may need small manual adjustments |
  114. ## Repository Map
  115. | Path | Purpose |
  116. |---|---|
  117. | [SKILL.md](./SKILL.md) | Internal Codex skill instructions |
  118. | [providers.md](./providers.md) | Provider-neutral API configuration notes |
  119. | [references/](./references/) | Figure templates, plotting rules, prompt templates, API notes |
  120. | [scripts/](./scripts/) | Image generation, plotting, setup checks, wizard scripts |
  121. | [examples/figure-briefs/](./examples/figure-briefs/) | Reusable figure brief examples |
  122. | [docs/examples/](./docs/examples/) | Showcase images, prompts, and notes |
  123. | [templates/figure-brief/](./templates/figure-brief/) | Platform-neutral figure brief templates |
  124. | [adapters/](./adapters/) | Claude Code and future platform adapters |
  125. ## What It Is Not
  126. - Not a full paper-upload web platform.
  127. - Not a replacement for checking scientific truth, labels, and numeric values.
  128. - Not a single prompt that treats plots, diagrams, and schematics as the same task.
  129. - Not a place to commit real API keys or private provider relay details.
  130. ## Notes
  131. - Keep real API keys outside the repository.
  132. - Prefer local plotting for exact quantitative figures.
  133. - Keep provider-specific private relay details out of public docs unless clearly marked as optional examples.
  134. - For exact publication plots, never rely on image generation for the values, axes, or benchmark geometry.
  135. ## License
  136. MIT License. See [LICENSE](./LICENSE).

Ratings and comments

Sign in to rate this skill.

Sign in

Free to use. If it’s helped, pass it on.