yjz211/hajimi-math-model-agent/legacy-modeling-plot-suite
Claude Code and agent skill by yjz211/hajimi-math-model-agent
- community
Checking that the SKILL.md file is still in the repository.
Manifest-driven router for mathematical-modeling competition figures. Reproduces the planning-to-render flow, routes each planned figure to the correct renderer, reconciles missing outputs, and audits final LaTeX size and page appearance.
Install
npx vskill@latest install yjz211/hajimi-math-model-agent/legacy-modeling-plot-suiteSKILL.md
Modeling Plot Suite
Use the pattern: classify upstream, record a machine-readable contract, execute the core render passes, then reconcile and audit. Do not choose a renderer from visual taste alone and do not load every drawing workflow.
Required decision sequence
- Read
references/router-contract.mdcompletely. For a full problem, paper, or multi-figure request, create or update the anchoredFIGURE_MANIFESTbefore drawing. Validate it withpython scripts/validate_figure_manifest.py --manifest <planning-document> --profile modeling --full-paper; fix every failure before rendering. For one explicit figure, apply the same classification rules; a separate manifest is optional. - Run
python scripts/bootstrap.py --workspace <active-project-root> --profile modeling-competition --capability allbefore the first deterministic figure. - Execute only the non-empty renderer classes, in this order:
- DATA/TABLE ->
paper-figure; - DETERMINISTIC DIAGRAMS ->
paper-technical-diagram, the manifest-compatible umbrella for Draw.io and TikZ, including engineering, physical, geometric, network, scheduling, architecture, and process diagrams; - ILLUSTRATION ->
paper-illustrationonly for qualifying pictorial or 3-D physical/engineering scenes; - HTML -> only when the user explicitly selects HTML instead of Draw.io;
- MERMAID -> only when Mermaid or text-native maintainability is requested.
- For each non-empty class, read
workflows/<name>.md, then its linkedreferences/<name>.md, completely and execute it without weakening its drawing prompt or review loop. - Reconcile the manifest after every pass. A missing planned output keeps the task open; retry or use only the documented fallback for that class. Never silently reclassify a missing figure.
- For every paper-bound figure, read
references/paper-layout-gate.mdand complete its inclusion-size and compiled-page checks before declaring completion.
When uncertain whether something is a data curve or a TikZ construction, classify it as DATA. Route all non-data diagrams that require exact, reproducible structure to paper-technical-diagram, then choose Draw.io or TikZ internally. AI illustration must not replace evidence plots, engineering schematics that can be drawn deterministically, precise topology, or mathematical geometry.
Completion gate
Finish only when the manifest is reconciled, every selected workflow passes its own checks, figures/latex_includes.tex has normalized sizes when LaTeX delivery is in scope, the paper uses those sizes unchanged, and the compiled paper page has been visually checked when a paper source is available.
Security
- Security scanPassed, 90 of 100Pattern checks on the skill's files.
- Model reviewPassed
Quality evaluation
- Task evaluationNot runNo evaluation yet. Run one in Skill Studio.
Source
- SKILL.md in the repositoryCheckingtoolkit/legacy-modeling-plot-suite/SKILL.md
Popularity
- 7-day trend score
- 0
Not refreshed yet.
Works with
Any agent vskill installs to: 56 targets in all. These 39 have a page on this site.
- Claude Code
- Cursor
- GitHub Copilot
- Windsurf
- Codex
- Gemini CLI
- Cline
- Amp
- Roo Code
- Goose
- Continue
- Kilo Code
- Junie
- Trae
- OpenHands
- Qwen Code
- AI Agent
- Tabnine Chat
- Sourcegraph Cody
- Amazon Q
- Firebase Genkit
- Sweep
- Devin
- Phind
- Mentat
- Claude Desktop
- Aider
- GPT Pilot
- Plandex
- bolt.new
- v0
- Replit Agent
- Anse
- TypingMind
- BetterChatGPT
- LibreChat
- Warp AI
- Pieces
- Fig AI
Embed badge
[](https://verified-skill.com/skills/yjz211/hajimi-math-model-agent/legacy-modeling-plot-suite)