detecting-performance-regressions

Compare current benchmark results against historical baselines to spot performance regressions. Use as an explicit/manual helper for build-to-build degradation review, not for broad optimization strategy or low-level profiling ownership.

Quality

63%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide

Securityby

Passed

No findings from the security scan

Fix and improve this skill with Tessl

tessl review fix ./bundled/skills/detecting-performance-regressions/SKILL.md

Quality

Content

37%

Reviews the quality of instructions and guidance provided to agents. Good implementation is clear, handles edge cases, and produces reliable results.

The SKILL.md body is commendably concise and well-organized as a positioning/overview document, but it is almost entirely descriptive: it provides no executable guidance, no sequenced workflow, and—most critically—never points Claude to the bundled scripts that implement the actual regression-detection work.

Suggestions

Add a workflow section with concrete, ordered steps that invoke the bundled scripts, e.g. 1) run scripts/analyze_metrics.py against the baseline and current results, 2) review the returned regression JSON, 3) run scripts/generate_report.py, 4) run scripts/create_github_comment.py to post the summary.

Reference the bundled materials explicitly from the body (e.g. a "## Tooling" section linking scripts/analyze_metrics.py and the references/ guides) so Claude can discover and use the executable assets rather than relying on the descriptive "Typical Outputs" list.

Include a validation checkpoint in the workflow — e.g. confirm analyze_metrics.py reports no regressions or that severity thresholds are met before generating and posting the report — so the multi-step batch operation has an explicit feedback loop.

Dimension	Reasoning	Score
Conciseness	The body is lean and assumes Claude's competence — it never explains what benchmarks, regressions, or CI/CD pipelines are, and every section (Positioning, When to Use, Not For, Typical Outputs, Related Skills) earns its place, matching the score-3 "every token earns its place" anchor.	3 / 3
Actionability	The body offers no concrete code or commands — "Typical Outputs" ("A baseline-vs-current regression summary", "Severity-ranked regressed metrics") describes results rather than instructing how to produce them, and it never references the bundled executable scripts (analyze_metrics.py, generate_report.py, create_github_comment.py), fitting the score-1 "describes rather than instructs" anchor.	1 / 3
Workflow Clarity	There is no sequenced procedure for the inherently multi-step analyze→report→comment pipeline, and "When to Use" lists use-cases rather than ordered steps, with no validation checkpoints, matching the score-1 anchor of steps missing and no validation mentioned; it is not a 2 because no steps are listed at all.	1 / 3
Progressive Disclosure	The body is well-organized into clear sections (not a monolithic wall), but it fails to signal or navigate to the real bundle files — the scripts that perform the actual work are present in ./scripts/ yet never referenced from SKILL.md, fitting the score-2 anchor of references present but not clearly signaled.	2 / 3
	Total	7 / 12 Passed

Description

90%

Based on the skill's description, can an agent find and select it at the right time? Clear, specific descriptions lead to better discovery.

The description is well-crafted: it states a concrete capability, supplies an explicit use-when trigger, and sharply distinguishes itself from neighboring skills via a negative-boundary clause. Its only weakness is that it names a single coupled action rather than enumerating multiple concrete actions.

Dimension	Reasoning	Score
Specificity	It names the domain and one concrete action — "Compare current benchmark results against historical baselines to spot performance regressions" — but this is a single coupled capability rather than the multiple distinct actions the score-3 anchor ("extract text, fill forms, merge documents") requires, so it is not comprehensive.	2 / 3
Completeness	It answers both what ("Compare current benchmark results against historical baselines to spot performance regressions") and when via an explicit trigger clause ("Use as an explicit/manual helper for build-to-build degradation review"), satisfying the score-3 anchor; it is not capped at 2 because the trigger guidance is present, not merely implied.	3 / 3
Trigger Term Quality	Natural terms a user would actually say are well covered: "benchmark results", "historical baselines", "performance regressions", and "degradation", matching the score-3 anchor for good coverage of natural terms.	3 / 3
Distinctiveness Conflict Risk	The explicit negative boundary ("not for broad optimization strategy or low-level profiling ownership") carves out a clear niche and reduces conflict with adjacent skills, matching the score-3 anchor for a clear niche with distinct triggers.	3 / 3
	Total	11 / 12 Passed

Validation

87%

Warnings & errors only

Checks the skill against the spec for correct structure and formatting. All validation checks must pass before discovery and implementation can be scored.

Validation — 14 / 16 Passed

Validation for skill structure

Criteria	Description	Result
allowed_tools_field	'allowed-tools' contains unusual tool name(s)	Warning
frontmatter_unknown_keys	Unknown frontmatter key(s) found; consider removing or moving to metadata	Warning

	Total	14 / 16 Passed

Repository: foryourhealth111-pixel/Vibe-Skills
Path: bundled/skills/detecting-performance-regressions/SKILL.md
Commit: f627ab5

Reviewed: about 9 hours ago

Table of Contents

Discovery Implementation Validation

Is this your skill?

If you maintain this skill, you can claim it as your own. Once claimed, you can manage eval scenarios, bundle related skills, attach documentation or rules, and ensure cross-agent compatibility.