feat: improve skill scores for GeneralizedNotationNotation

[yogesh-tessl]

Hey @docxology 👋

A 25-step processing pipeline for formalising Active Inference generative models into standardised text notation. The combination of DOI-registered research rigor with practical CLAUDE.md integration shows you're serious about making this both academically sound and agent-usable. Wanted to suggest a few improvements to the SKILL.md.

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
gnn-api	29%	95%	+66%
gnn-cli-dispatch	69%	83%	+14%
gnn-pipeline	89%	100%	+11%
gnn-parsing	89%	94%	+5%
gnn-mcp-protocol	79%	84%	+5%

Changes Summary

• gnn-api - Rewrote from stub to full skill with proper description, concrete API examples, HTTP workflow with curl commands, and references
• gnn-cli-dispatch - Added YAML frontmatter with description and "Use when" clause, replaced comment-only programmatic section with a recommended workflow, added references
• gnn-pipeline - Trimmed redundant intro sections, added validation checkpoints between pipeline phases with error recovery guidance
• gnn-parsing - Added a recommended discover/parse/validate workflow with error handling, removed duplicated Documentation section
• gnn-mcp-protocol - Expanded description with more concrete trigger terms, added recommended workflow with tool registration verification, removed duplicated Documentation section

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 (https://docs.tessl.io/evaluate/optimize-a-skill-using-best-practices) and ask it to optimize your skill. Ping me - @yogesh-tessl - if you hit any snags.

Thanks in advance 🙏

[docxology]

Thanks for the contribution. I can't approve this as-is because it weakens the top-level skill contract by removing the explicit 'When to Use This Skill' routing block, and it adds at least one stale validation checkpoint path: the pipeline uses unpadded output directories such as output/3_gnn_output, not output/03_gnn_output/parsed. Please keep the trigger/routing guidance intact and verify every added command/path against the current repository before this lands.