fix: improve generate-smoke-test skill score (85% → 92%)#75
Merged
gjtorikian merged 2 commits intoMay 21, 2026
Merged
Conversation
Hey @gjtorikian 👋 I ran your skills through `tessl skill review` at work and found some targeted improvements for `generate-smoke-test`. Here's the full before/after: | Skill | Before | After | Change | |-------|--------|-------|--------| | generate-smoke-test | 85% | 92% | +7% | | review-operations | 100% | 100% | — | | generate-emitter | 93% | 93% | — | | check-emitter-parity | 90% | 90% | — | | generate-sdk | 90% | 90% | — | | verify-compat | 90% | 90% | — | | generate-extractor | 90% | 90% | — | | verify-smoke-test | 90% | 90% | — | | integrate | 85% | 85% | — | <details> <summary>Changes made to <code>generate-smoke-test</code></summary> - **Consolidated operations map references** — the `.oagen-manifest.json` explanation was repeated across Steps 2, 3, 5, and the Emitter-Fixing Loop. Kept the authoritative explanation in Step 2 and replaced the others with concise back-references. - **Replaced interception prose with a table** — Step 1 listed language-specific interception libraries in prose format. A 5-row table conveys the same information more efficiently. - **Extracted implementation patterns to a reference file** — moved the concrete `buildArgs()` template and full smoke script skeleton to `references/implementation-patterns.md`, improving progressive disclosure while keeping the main SKILL.md focused on workflow. - **Trimmed redundant explanatory text** — removed a self-describing sentence in the overview that restated what the structure already communicates. </details> I also stress-tested your `generate-emitter` skill against a few real-world task evals and it held up really well on scaffolding a Go emitter with full `SdkBehavior` integration and operations map support. 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. 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](https://github.com/yogesh-tessl) — if you hit any snags. Thanks in advance 🙏
gjtorikian
changed the title
Collaborator
|
great, thanks! |
Contributor
Author
|
@gjtorikian, noticed you renamed the prefix to fix: and pushed a formatting commit before merging, which tells me you have release-please or something similar keyed off conventional commits. That kind of pipeline discipline across 9 skills with a release bot cutting 0.21.0 automatically is a clean setup. Glad this landed well. I’ve reviewed a bunch of skills now and have picked up a few useful patterns around keeping them maintainable and secure as usage grows. Happy to swap notes over a quick 15-min coffee and share the latest practices we've seen. also eager to hear how you’re thinking about use cases on your side, as it helps us refine the approach. |
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
2 participants
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Hey @gjtorikian 👋
the way you've structured the skills around the SDK generation lifecycle is really well thought out. Having separate skills for emitter generation, extraction, smoke testing, and compatibility checks maps cleanly to the actual workflow of maintaining a generated SDK.
ran your skills through
tessl skill reviewat work and found some targeted improvements forgenerate-smoke-test. Here's the full before/after:Changes made to
generate-smoke-test.oagen-manifest.jsonexplanation was repeated across Steps 2, 3, 5, and the Emitter-Fixing Loop. Kept the authoritative explanation in Step 2 and replaced the others with concise back-references.buildArgs()template and full smoke script skeleton toreferences/implementation-patterns.md, improving progressive disclosure while keeping the main SKILL.md focused on workflow.quick honest disclosure. I work at https://github.com/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 @yogesh-tessl, if you hit any snags.