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

62%

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

51%Weight 40%Scale 1-5

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

The body is admirably concise and well-organized as a positioning document, but it functions as a skill overview rather than actionable instructions: it never points to the bundled scripts or gives a runnable workflow with validation. It needs concrete execution guidance to move beyond a positioning stub.

Suggestions

Add a 'How to use' workflow that invokes the bundled scripts in order, e.g. `python scripts/analyze_metrics.py <metrics> --baseline <baseline>` then `generate_report.py` then `create_github_comment.py`, with an explicit validation step (check the regression JSON before posting).

Link the bundled scripts from the body (e.g. "See [scripts/analyze_metrics.py](scripts/analyze_metrics.py)") so the real bundle files are clearly signaled and progressive disclosure improves.

Provide one concrete example invocation with expected output and threshold/severity parameters so the guidance is copy-paste ready rather than descriptive.

Dimension	Reasoning	Score
Conciseness	The body is lean with short sections ("Positioning", "When to Use", "Not For / Boundaries", "Typical Outputs", "Related Skills") and no padding or explanation of concepts Claude already knows. It matches the 5 anchor where every token earns its place.	5 / 5
Actionability	The body gives high-level hints ("baseline-vs-current regression summary", "Severity-ranked regressed metrics") but no concrete commands, thresholds, or invocations of the bundled scripts (e.g. analyze_metrics.py). It describes rather than instructs, sitting above the 1 anchor (purely vague) but below the 3 anchor (some concrete guidance).	2 / 5
Workflow Clarity	"Typical Outputs" lists deliverables rather than a sequenced workflow, and there are no steps tying the bundled scripts together nor any validation checkpoint. This matches the 2 anchor (rough sequence present but many gaps, validation absent) rather than the 3 anchor (steps listed with validation gaps), since no real steps are enumerated.	2 / 5
Progressive Disclosure	The body is well-sectioned and the bundle has real script files (analyze_metrics.py, generate_report.py, create_github_comment.py), but the body never references or links to them, and references/assets are only placeholder READMEs. It sits at the 3 anchor (some structure, references not clearly signaled) rather than 4 (references mostly clear) because the scripts are not surfaced.	3 / 5
	Total	12 / 20 Passed

Description

73%Weight 40%Scale 1-5

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 concise, third-person, and clearly scoped with explicit boundary guidance, giving it strong distinctiveness and completeness. Its main weakness is limited action specificity and a few missing natural trigger synonyms.

Suggestions

Expand the action list beyond 'compare' and 'spot' to concrete verbs like 'analyze benchmark deltas', 'rank regressed metrics by severity', and 'report degradation' to lift specificity toward 5.

Add a more user-facing trigger clause such as 'Use when the user mentions benchmarks, performance regressions, or build-to-build slowdowns' to make the 'when' more explicit and add natural synonyms.

Include common shorthand terms like 'benchmarks' and 'perf regression' alongside the formal phrases to round out trigger term coverage.

Dimension	Reasoning	Score
Specificity	Quotes "Compare current benchmark results against historical baselines to spot performance regressions" name the domain and one to two concrete actions (compare against baselines, spot regressions) but offer no comprehensive action list. It sits below 4 (which needs several specific actions with only minor gaps) and above 2 (which barely names actions).	3 / 5
Completeness	It answers "what" (compare benchmarks vs baselines to spot regressions) and gives an explicit "when" via "Use as an explicit/manual helper for build-to-build degradation review". The "when" is present but not as concrete/trigger-rich as the 5 anchor, and it is more explicit than the 3 anchor which lacks "when".	4 / 5
Trigger Term Quality	Phrases "benchmark results", "historical baselines", "performance regressions", "build-to-build degradation review" give good natural keyword coverage. It misses some common synonyms (e.g. "benchmarks", "perf regression"), so it is not the comprehensive 5 but is clearly above the 3 anchor.	4 / 5
Distinctiveness Conflict Risk	The boundary clause "not for broad optimization strategy or low-level profiling ownership" carves a clear niche with distinct triggers and minimal conflict risk, matching the 5 anchor. It is more differentiated than the 4 anchor's minor overlap risk.	5 / 5
	Total	16 / 20 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: d5ae560

Reviewed: 1 day 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.