Mind Map

Generate a PNG mind map from Markdown using free (auto), center, or horizontal layout. Use when user asks to produce a brain map/思维导图 from notes.

Security Scan

The skill appears to do what it claims (generate PNG mind maps from Markdown); its requirements and runtime behavior are coherent and proportionate.

Purpose & Capability

Name/description align with included code: scripts parse Markdown and render PNG images using Pillow. Declared runtime artifacts (examples, reference, runner) match the functionality. No unrelated credentials, binaries, or config paths are requested.

Instruction Scope

Runtime instructions stay within the stated purpose (read Markdown input, optionally from a file, generate image(s), save outputs and report stats). Two minor issues: SKILL.md includes a hardcoded Windows-oriented working directory example ('D:\Work\skills\mind_map') and there are inconsistent path/name conventions in docs (uses both mind_map and mind-map-skill), which are usability inconsistencies but not security-critical. Also note the runner will read any file path you pass and will save the input Markdown into the output folder (so do not pass sensitive files you do not want duplicated into output).

Install Mechanism

This is instruction/code bundle with no automated install spec. Dependencies are minimal (Pillow listed in requirements.txt). The only network activity is an on-demand font download from a GitHub raw URL (FONT_DOWNLOAD_URL) to a local cache; using raw.githubusercontent.com is common for assets but still an external network fetch. There is no download-from-arbitrary-URL installer or obfuscated fetches.

Credentials

The skill requests no environment variables or credentials. The only filesystem writes are to the output directory and to a per-user cache (~/.cache/mind-map-skill/chinese_font.ttc) for an optional font — this is proportionate to its purpose.

Persistence & Privilege

always is false and disable-model-invocation is true (so it will not be autonomously invoked). The skill writes only to its own output and cache paths and does not alter other skills or system-wide agent settings.

Assessment

This skill is coherent and appears safe for typical use: it parses Markdown and renders PNGs locally. Before installing or running, note: (1) you need Pillow installed (requirements.txt lists pillow>=10); (2) on first run it will try to download a CJK font from raw.githubusercontent.com and cache it at ~/.cache/mind-map-skill/chinese_font.ttc — if you prefer no network fetch, either provide a local CJK font or run offline; (3) the runner will read any file path you pass and will save the input Markdown into the output folder, so avoid pointing it at sensitive files you don't want duplicated; (4) SKILL.md contains inconsistent example paths (Windows path and different directory names) — ensure you call the correct script path for your environment. If you need higher assurance, review the bundled Python files locally (they are small and readable) and run the tool in an isolated environment (container or VM).

Mind Map Skill

Generate mind map images locally from Markdown content.

Invocation

Use this skill directly:

• /mind-map-skill free docs/plan.md
• /mind-map-skill center "# 项目\n## 模块\n- A\n- B"
• /mind-map-skill horizontal notes.md

Arguments are mapped as:

• $0: layout type (free, center, horizontal)
• $1: markdown file path or inline markdown text

If layout argument is missing, use free.

Execution steps

1. Parse $ARGUMENTS.
2. Resolve markdown input:
   • If $1 is a file path, read file content.
   • Otherwise treat $1 as inline markdown.
3. Run from repository root (D:\Work\skills\mind_map):

python mind-map-skill/scripts/run_mind_map.py --layout <layout> --input "<file-or-markdown>" --output-dir mind_map_output

By default, output files are grouped by date under mind_map_output/YYYY-MM-DD/. Use --flat-output to write directly into mind_map_output/.

4. Report:
   • selected layout
   • output PNG path
   • input markdown saved path (same dated folder)
   • node count and depth

Layout behavior

• center: radial structure (equivalent to mind_map_center capability)
• horizontal: left-to-right structure (equivalent to mind_map_horizontal capability)
• free: auto select based on complexity (equivalent to mind_map_free capability)

Auto rule in free mode:

• choose center when depth <= 4 and total_nodes <= 100
• otherwise choose horizontal

Supporting files

• Runtime engine: mind-map-skill/scripts/mind_map_generator.py
• Runner script: mind-map-skill/scripts/run_mind_map.py
• Example markdown: mind-map-skill/examples/sample.md
• Parameter schema (optional compatibility): mind-map-skill/skill.yaml
• Detailed reference: mind-map-skill/reference.md
• Usage examples: mind-map-skill/examples.md

Font behavior

• On first run, the skill attempts to download a CJK font to local cache: ~/.cache/mind-map-skill/chinese_font.ttc.
• Download source in repository: resources/chinese_font.ttc (not part of skill upload package).
• If download is unavailable, it falls back to prioritized system CJK fonts (Linux/macOS/Windows specific order).

Project metadata

• Author: @sawyer-shi
• Email: [e-mail]