performance

Profiling methodology and optimization strategy for performance work. Use when the user asks to "make this faster", "optimize", "profile", "reduce latency", "fix slow", "improve throughput", or when investigating performance regressions.

Quality

—

Does it follow best practices?

Impact

—

No eval scenarios have been run

Securityby

Passed

No known issues

Quality

Content

85%

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

A well-structured methodology skill: a lean, dense overview with a clear measured loop, trade-off and bottleneck frameworks, and a properly signaled one-level reference holding the executable commands. The main gap is that the body itself carries no executable profiler commands, so its standalone actionability is framework-level.

Suggestions

Add one or two inline copy-paste profiler commands (e.g. a canonical CPU-profile snippet) in the Profiling Strategy section so the body is actionable without opening the reference.

Make the loop's error-recovery branch explicit in the body ("If no improvement: revert and return to step 3 with a new hypothesis") rather than only in the reference checklist.

Consider trimming or relocating the Knuth quote to the reference so every remaining line in the overview is directly actionable.

Dimension	Reasoning	Score
Conciseness	The body is dense and non-redundant for the breadth it covers — compact trade-off and tool-reveal tables, terse bottleneck pattern entries — and assumes Claude's competence without explaining what profiling or latency is. The only mild decoration is the brief Knuth quote, which still reinforces a real principle. Not 2 because there is no padding or basic-concept explanation to trim.	3 / 3
Actionability	Concrete framework guidance is present (specific remedies like "Batch operations, add caching, use async I/O", the output-format template, "use at least two tool types"), but the actual executable profiler commands are deferred to the reference rather than appearing inline. Not 1 because the guidance is specific, not vague; not 3 because the body itself is not copy-paste executable.	2 / 3
Workflow Clarity	The Performance Loop is a clear 6-step numbered sequence with an explicit validation checkpoint ("Measure again — Did it help? By how much? Any regressions elsewhere?") and a guardrail ("Never skip from step 1 to step 5"). Not 2 because the checkpoint is explicit, not implicit.	3 / 3
Progressive Disclosure	The body is an overview that points to a single, clearly signaled, one-level-deep reference ("see references/profiling-checklist.md"), and that file exists and does not nest further. Content is appropriately split: methodology/framework inline, concrete commands in the reference. Not 2 because the reference is well-signaled and genuinely one level deep.	3 / 3
	Total	11 / 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.

A strong description: it answers both what and when with an explicit, natural-language trigger list targeting a clear performance niche. The only weakness is that the capability statement is somewhat abstract ("methodology and strategy") rather than enumerating concrete actions.

Suggestions

Replace the abstract "Profiling methodology and optimization strategy" opener with concrete actions, e.g. "Profile bottlenecks, measure baselines, and apply targeted optimizations for performance work."

Consider adding a couple of negative-boundary triggers (e.g. "Not for algorithm design or capacity planning") to further reduce conflict risk.

Dimension	Reasoning	Score
Specificity	"Profiling methodology and optimization strategy for performance work" names the domain and approach but describes actions abstractly rather than listing multiple concrete capabilities. Not 3 because it lacks several specific concrete actions; not 1 because the domain and core activities (profiling, optimizing) are named.	2 / 3
Completeness	It states what the skill does ("Profiling methodology and optimization strategy for performance work") and explicitly when to use it ("Use when the user asks to..."). Both what and when are present with explicit triggers.	3 / 3
Trigger Term Quality	The trigger list — "make this faster", "optimize", "profile", "reduce latency", "fix slow", "improve throughput", "investigating performance regressions" — is broad natural language users would actually say. Matches the anchor for good coverage of natural terms.	3 / 3
Distinctiveness Conflict Risk	The performance-profiling niche is clear and the triggers (latency, throughput, profile, regressions) are distinct from other skills, making false triggers unlikely. Not 2 because the combination of terms pins it to a specific domain.	3 / 3
	Total	11 / 12 Passed

Validation

100%

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 — 16 / 16 Passed

Validation for skill structure

No warnings or errors.

Repository: tslateman/duet
Commit: 436824a

Reviewed: 5 days 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.