feat: optimize cls-certify skill score from 71% to 94%

[rohan-tessl]

Hey @CatREFuse 👋

I ran your skills through tessl skill review at work and found some targeted improvements. Here's the full before/after:

Skill	Before	After	Change
cls-certify	71%	94%	+23%

Changes made

SKILL.md (~72% size reduction, 1069 → 294 lines)

• Restructured frontmatter: Moved version, build, batch_mode, output_dir, scan_mode into metadata block with proper string values — resolves all 3 validation warnings
• Extracted scoring matrix: Created references/scoring-matrix.md with the full deduction table, rating standards, intent verification rules, and agent context injection guidelines — previously inline across multiple sections
• Condensed six-dimension detection: Replaced verbose inline regex patterns, concept explanations, and detection examples with concise summaries pointing to existing reference files (threat-patterns.md, sensitive-data-patterns.md, api-classification.md, etc.)
• Added validation checkpoints: Explicit "验证点" gates between phases (verify JSON outputs exist, confirm classification tier, check score results before continuing)
• Removed redundant content: Eliminated inline explanations of well-known concepts (SQL injection, eval(), GDPR requirements) that the LLM already understands
• Consolidated HTML rendering details: Replaced inline placeholder injection rules and SVG coordinate calculations with a pointer to references/report-data-protocol.md
• Preserved all workflow functionality: Every phase, bash command, tool invocation, and decision point is intact — just more concise and actionable

Honest disclosure — I work at @tesslio where we build tooling around skills like these. Not a pitch - just saw room for improvement and wanted to contribute.

Want to self-improve your skills? Just point your agent (Claude Code, Codex, etc.) at this Tessl guide and ask it to optimize your skill. Ping me - @rohan-tessl - if you hit any snags.

Thanks in advance 🙏