feat: improve skill review scores (+51% avg)

[rohan-tessl]

Hey @dtsong 👋

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
language-conventions	38%	92%	+54%
dockerfile-generation	42%	97%	+55%
git-workflows	42%	97%	+55%
terraform-skill	43%	97%	+54%
workflow	43%	83%	+40%

This PR is intentionally scoped to the five lowest-scoring skills to keep it reviewable. The remaining seven skills can be improved in follow-ups or via automated review on future PRs.

Changes summary

Across all five skills, the improvements follow a consistent pattern:

• Expanded frontmatter descriptions with explicit "Use when..." clauses and natural trigger terms so agents can reliably select the right skill
• Added concrete, executable code examples — bash commands, Dockerfile snippets, HCL blocks, git commands — replacing abstract process descriptions
• Structured workflows with validation checkpoints — each step now includes specific commands to run and verify before moving to the next
• Improved quality checklists with more specific, actionable items tied to the concrete examples in each skill
• Preserved the existing section structure (Purpose, Inputs, Process, Output Format, Quality Checks) per your CONTRIBUTING.md and AGENTS.md conventions

Specific per-skill highlights:

• language-conventions: Added config file detection commands for Python/TypeScript/Terraform and concrete validation commands (ruff check, npx tsc --noEmit, terraform fmt)
• dockerfile-generation: Added a complete multi-stage Dockerfile example with non-root user, healthcheck, and build/verify commands
• git-workflows: Added explicit commands for each workflow step including branch naming conventions, rebase conflict recovery, and safe cleanup
• terraform-skill: Added HCL code examples with variable validation, full fmt/validate/plan sequence, and security scanning commands
• workflow: Added quality gate detection scripts, multi-language test runner examples, and test-after-each-commit feedback loops

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 🙏