一键在 Manus 中运行任何 Skill

markdown-mermaid-support

星标4

分支1

更新时间2026年3月1日 18:04

Create or revise technical Markdown documents that need Mermaid diagrams, flow validation, and source-code evidence links. Use when docs describe workflows/dependencies/branches and should include clickable file:line references plus concrete code snippets.

安装

用 Codex 或 Claude 帮你安装复制这段 Prompt，粘贴到 Codex、Claude 或其他助手里，让它检查 Skill 页面并帮你完成安装。

在 Manus 中运行

来源

CHDTevior

CHDTevior/Tengjiao_codex_skill

打开 GitHub 仓库查看创作者相关仓库

下载

在 Manus 中运行

Markdown Mermaid Support

Use this skill to produce high-clarity technical Markdown with two outcomes:

Correct Mermaid diagrams for process and dependency logic.
Clickable code references with concrete source snippets.

Workflow

Identify sections that need diagrams.

Prefer Mermaid when logic has 3+ steps, branching, or module handoffs.
Use flowchart for process/dependency, sequenceDiagram for interaction timelines.

Validate diagram semantics before writing.

Verify node order matches real execution order.
Explicitly show optional branches and gate conditions.
Do not mix training loop and inference loop in one path unless clearly separated.

Add/refresh code evidence links.

Convert path/to/file.py:123 to clickable markdown links.
Keep references near the owning sentence/table.
Add a code snippet section for referenced lines.

Run the bundled enhancer script when editing docs.

scripts/md_code_ref_enhancer.py \
  --repo-root <repo-root> \
  --files <doc1.md> [<doc2.md> ...] \
  --write \
  --append-snippets

Re-check rendering.

Mermaid blocks must render in Markdown preview.
Links must jump to the target file and line.
Snippets must match referenced lines.

Authoring Rules

Keep diagrams local to the section they explain.
Keep each diagram focused on one question.
Keep labels short and action-oriented.
Avoid oversized diagrams; split by concern.
Prefer idempotent doc updates (re-running script should not duplicate links/snippets).

References

Mermaid templates: references/mermaid-patterns.md
Code evidence patterns: references/code-evidence-patterns.md

同仓库更多 Skills

同仓库

genai-evaluation-metrics

CHDTevior/Tengjiao_codex_skill

Use when evaluating generative models — choosing metrics (FID, IS, KID, sFID, FDD, FVD, PRDC, LPIPS, SSIM, AuthPct, Vendi), setting up online or offline evaluation, feature extractor selection, distributed computation, memory management during sampling.

2026-04-144

gpu-training-acceleration

CHDTevior/Tengjiao_codex_skill

Use when optimizing PyTorch training speed or memory on CUDA GPUs — global flags, torch.compile, fused optimizers, mixed precision, gradient checkpointing, kernel fusion, memory layout, or latent-space training. Applies to any PyTorch training workload.

2026-04-144

hf-dataset-management

CHDTevior/Tengjiao_codex_skill

Use when curating, uploading, or managing HuggingFace datasets for ML training, including offline caching, preflight verification, and data directory conventions.

2026-04-144

hydra-experiment-config

CHDTevior/Tengjiao_codex_skill

Use when structuring ML experiment configs with Hydra, adding new config groups, or debugging config resolution. Applies to any project using Hydra for hyperparameter management.

2026-04-144

ml-ablation-design

CHDTevior/Tengjiao_codex_skill

Use when designing ablation studies to compare model components, loss functions, or architectural choices. Covers synthetic data experiments, variant loops, production metrics, and W&B grouping.

2026-04-144

wandb-experiment-tracking

CHDTevior/Tengjiao_codex_skill

Use when integrating W&B experiment tracking into ML training pipelines, including logging strategy, run configuration, and online/offline mode management.

2026-04-144

name	markdown-mermaid-support
description	Create or revise technical Markdown documents that need Mermaid diagrams, flow validation, and source-code evidence links. Use when docs describe workflows/dependencies/branches and should include clickable file:line references plus concrete code snippets.

Markdown Mermaid Support

Use this skill to produce high-clarity technical Markdown with two outcomes:

Correct Mermaid diagrams for process and dependency logic.
Clickable code references with concrete source snippets.

Workflow

Identify sections that need diagrams.

Prefer Mermaid when logic has 3+ steps, branching, or module handoffs.
Use flowchart for process/dependency, sequenceDiagram for interaction timelines.

Validate diagram semantics before writing.

Verify node order matches real execution order.
Explicitly show optional branches and gate conditions.
Do not mix training loop and inference loop in one path unless clearly separated.

Add/refresh code evidence links.

Convert path/to/file.py:123 to clickable markdown links.
Keep references near the owning sentence/table.
Add a code snippet section for referenced lines.

Run the bundled enhancer script when editing docs.

scripts/md_code_ref_enhancer.py \
  --repo-root <repo-root> \
  --files <doc1.md> [<doc2.md> ...] \
  --write \
  --append-snippets

Re-check rendering.

Mermaid blocks must render in Markdown preview.
Links must jump to the target file and line.
Snippets must match referenced lines.

Authoring Rules

Keep diagrams local to the section they explain.
Keep each diagram focused on one question.
Keep labels short and action-oriented.
Avoid oversized diagrams; split by concern.
Prefer idempotent doc updates (re-running script should not duplicate links/snippets).

References

Mermaid templates: references/mermaid-patterns.md
Code evidence patterns: references/code-evidence-patterns.md