performance

MUST be used when fixing Flows app performance — re-renders, query patterns, pagination, unbounded fetches, LLM-over-query-results, bundles, memory leaks. Measure before and after. Triggers: performance, slow, laggy, optimize, re-render, bundle size, CDF query, virtualize, chat completions, LLM cost.

Quality

88%

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

Quality

Content

77%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.

A highly actionable, well-sequenced performance workflow with strong before/after measurement validation. Its main weaknesses are length/repetition that could be trimmed and a monolithic structure with no progressive disclosure via reference files.

Suggestions

Collapse redundant Step 5 table rows (e.g., 'No limit set' and 'Missing limit parameter') and standardize the per-step 'apply the fix' phrasing to cut repetition and reduce length.

Offload the dense BAD/FIX reference tables and the DMS query rationale into a references/ file (e.g., references/dms-query-patterns.md) and link to it one level deep, keeping SKILL.md a lean overview.

Move the Step 9 bundle-analysis details and the visualizer setup/teardown into a script (scripts/analyze-bundle.sh) so the body stays a concise checklist.

Dimension	Reasoning	Score
Conciseness	Mostly efficient and dense with actionable code/commands, but at ~525 lines it carries structural repetition (e.g., duplicate Step 5 table rows 'No limit set' vs 'Missing limit parameter', recurring 'apply the fix directly' phrasings) that could be tightened.	3 / 5
Actionability	Fully executable guidance throughout: copy-paste grep commands and complete BAD/FIX code for useMemo/useCallback, instances.query rewrites, useInfiniteQuery, the @tanstack/react-virtual pattern, React.lazy/Suspense, and AbortController cleanup covering the common cases.	5 / 5
Workflow Clarity	Clear 11-step sequence with explicit validation bookends (Step 1 baseline mandate, Step 11 re-measure and report delta with 'Do not fabricate numbers') and per-step checklist tables; the measurement feedback loop satisfies validation for the batch/destructive fixes.	5 / 5
Progressive Disclosure	Well-sectioned into steps but entirely monolithic — no bundle files exist (references/, scripts/, assets/ absent) and a 525-line skill inlines reference-like detail (fix tables, semantic-knowledge rationale) that could be offloaded; the one reference to a semantic-knowledge/ directory is conditional and external to the bundle.	3 / 5
	Total	16 / 20 Passed

Description

100%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.

An exemplary description: concrete capabilities, explicit 'MUST be used when' trigger guidance, a dedicated natural-language triggers list, and a tightly scoped niche. Third-person/imperative voice with no over-claims.

Dimension	Reasoning	Score
Specificity	Lists multiple concrete issue categories ('re-renders, query patterns, pagination, unbounded fetches, LLM-over-query-results, bundles, memory leaks') plus the concrete action 'Measure before and after' — comprehensive coverage of the performance domain.	5 / 5
Completeness	Explicitly answers both what ('fixing Flows app performance … Measure before and after') and when ('MUST be used when fixing Flows app performance' plus a dedicated 'Triggers:' list) with concrete trigger phrases.	5 / 5
Trigger Term Quality	Comprehensive natural-term coverage including synonyms ('slow', 'laggy', 'optimize') alongside domain terms users say ('bundle size', 'virtualize', 'LLM cost'), matching the comprehensive-coverage anchor.	5 / 5
Distinctiveness Conflict Risk	Tight niche scoped to the Flows app, CDF/DMS query patterns, and LLM-over-query-results with distinct triggers, giving minimal conflict risk with other skills.	5 / 5
	Total	20 / 20 Passed

Validation

81%

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 — 13 / 16 Passed

Validation for skill structure

Criteria	Description	Result
skill_md_line_count	SKILL.md is long (534 lines); consider splitting into references/ and linking	Warning
allowed_tools_field	'allowed-tools' contains unusual tool name(s)	Warning
metadata_version	'metadata.version' is missing	Warning

	Total	13 / 16 Passed

Repository: cognitedata/builder-skills
Path: skills/performance/SKILL.md
Commit: c87160a

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.