Tessl • The management layer for AI agent skills

SkillPublic

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

79%

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 ./skills/performance/SKILL.md

Quality

Content

70%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 playbook with a real measure-before/measure-after validation loop and concrete BAD/FIX examples throughout. Its weaknesses are length and monolithic structure: generic React boilerplate and per-fix code templates are inlined in SKILL.md instead of split into reference files, and a few terse hard-gate lines (LLM completion ceiling, undefined agent resources) are ambiguous.

Suggestions

Move the generic React fix templates (virtualizer boilerplate, useDebouncedValue implementation, memory-leak cleanup examples) into a references/ file (e.g. references/react-fixes.md) and keep SKILL.md to the repo-specific greps, decision tables, and one compact example per fix.

Clarify the LLM hard-gate line: state units for "5 / ceiling 50" (e.g. 'max 5 concurrent completions, hard ceiling of 50 rows') and either define sendAgentMessage / integrate-fusion-agent inline or point to a reference that does.

De-duplicate Step 5's table ("No limit set" and "Missing limit parameter" are the same fix) and merge the overlapping staleTime/refetchOnWindowFocus greps in Step 6 to trim tokens.

Dimension	Reasoning	Score
Conciseness	Dense and repo-specific overall, but ~530 lines inline substantial boilerplate Claude already knows — a 40-line virtualizer template, a full useDebouncedValue implementation, three standard memory-leak cleanup examples — plus duplicated Step 5 table rows ("No limit set" / "Missing limit parameter") and overlapping Step 6 greps. Mostly efficient but clearly could be tightened, fitting anchor 3.	3 / 5
Actionability	Exact grep commands, copy-paste BAD/FIX code pairs, issue-to-fix tables, and install/uninstall commands make the guidance largely executable. Minor gaps keep it below 5: "If per-item completions remain: 5 / ceiling 50" is cryptic (units unstated), and sendAgentMessage / integrate-fusion-agent are referenced without definition.	4 / 5
Workflow Clarity	Clear Steps 1-11 sequence with a genuine measurement feedback loop: "Record baseline numbers. Every fix must be measured against these", Step 11 re-runs the audit and reports the delta, with "If a step produced no improvement, state that explicitly. Do not fabricate numbers." The tables function as per-step checklists, and the batch-fix workflow is explicitly validated, so the batch-cap does not apply.	5 / 5
Progressive Disclosure	No bundle files exist (no references/, scripts/, or assets/); everything is inlined in one ~530-line SKILL.md. Per-fix code templates (DMS query rewrites, virtualizer, debounce, leak cleanups) could live in separate reference files, and the only pointer — "consult the semantic-knowledge/ directory if available in the workspace" — is hedged and not a clearly signaled bundle file. Structure from 11 clean sections keeps it above anchor 2's no-structure case.	3 / 5
	Total	15 / 20 Passed

Description

83%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.

A strong description: it names a specific niche (Flows app performance), enumerates concrete problem areas, includes an explicit 'MUST be used when' clause and a dedicated trigger list. Main gaps are a few missing natural synonyms in the trigger list and slightly generic terms (performance, slow, optimize) that could collide with general perf skills.

Dimension	Reasoning	Score
Specificity	Enumerates several concrete capability areas ("re-renders, query patterns, pagination, unbounded fetches, LLM-over-query-results, bundles, memory leaks") plus the action "Measure before and after". Not a 5 because these are issue-areas rather than a comprehensive list of fix-actions; clearly above 3's 1-2 concrete items.	4 / 5
Completeness	Both questions answered explicitly: "MUST be used when fixing Flows app performance" states what and when, and a concrete "Triggers:" list of natural phrases follows. Matches the anchor-5 example structure; anchor 4 would require the 'when' to be less explicit than it is.	5 / 5
Trigger Term Quality	"Triggers: performance, slow, laggy, optimize, re-render, bundle size, CDF query, virtualize, chat completions, LLM cost" gives good natural-term coverage. Missing a few obvious synonyms and extensions ("memory leak", "profiler", "fetch all"), so it sits at anchor 4 rather than 5.	4 / 5
Distinctiveness Conflict Risk	Scoped to the "Flows app" with CDF/DMS-specific terms ("CDF query", "LLM-over-query-results") giving a clear niche, but generic triggers like "performance", "slow", and "optimize" create minor overlap risk with general optimization skills — anchor 4, not 5.	4 / 5
	Total	17 / 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
Revision: 7be219f

Reviewed: 1 month ago

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.

performance

Table of Contents

Is this your skill?