Skip to content

feat: improve 5 lowest-scoring skill definitions#66

Open
rohan-tessl wants to merge 1 commit intodatabricks:mainfrom
rohan-tessl:improve/skill-review-optimization
Open

feat: improve 5 lowest-scoring skill definitions#66
rohan-tessl wants to merge 1 commit intodatabricks:mainfrom
rohan-tessl:improve/skill-review-optimization

Conversation

@rohan-tessl
Copy link
Copy Markdown

Hey @arsenyinfo 👋

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

Skill Before After Change
databricks-dabs 71% 100% +29%
databricks-core 76% 100% +24%
databricks-jobs 81% 97% +16%
databricks-pipelines 89% 94% +5%
databricks-lakebase 89% 94% +5%

These were easy changes to bring the skill's structure and activation in line with what performs well against Anthropic's best practices.

What changed in databricks-dabs

Added quoted description with explicit "Use when" clause. Added compatibility and metadata.version fields. Removed redundant "When to Use This Skill" body section. Added a concrete quick-start workflow (init, configure, validate, deploy, run) with a minimal databricks.yml example.

What changed in databricks-core

Replaced vague description with concrete action verbs and explicit "Use when" clause listing specific trigger scenarios (CLI commands, authentication setup, workspace configuration, bundle deployment, data exploration).

What changed in databricks-jobs, databricks-pipelines, databricks-lakebase

Moved non-standard parent frontmatter key into metadata.parent. Expanded descriptions with concrete actions and "Use when" clauses. Removed verbose inline CLAUDE.md/AGENTS.md templates. Consolidated examples where content was duplicated.

In addition, I stress-tested your databricks-dabs skill against a few real-world scenarios, and it held up really well. This means that your skill meaningfully improves agent steering and contributes to stronger output quality. Kudos for that!

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.

If you want to self-improve your skills, or define your own scenarios to pressure test, just ask your agent (Claude Code, Codex, etc.) to evaluate and optimize your skill with Tessl. Ping me @rohan-tessl, if you hit any snags.

@arsenyinfo
Copy link
Copy Markdown
Collaborator

@camielstee-db I think it needs your review

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

2 participants