agent-performance-benchmarker
Agent skill for performance-benchmarker - invoke with $agent-performance-benchmarker
31
0%
Does it follow best practices?
Impact
81%
2.89xAverage score across 3 eval scenarios
Passed
No known issues
Optimize this skill with Tessl
npx tessl skill review --optimize ./.agents/skills/agent-performance-benchmarker/SKILL.mdQuality
Discovery
0%Based on the skill's description, can an agent find and select it at the right time? Clear, specific descriptions lead to better discovery.
This is an extremely weak description that provides virtually no useful information for skill selection. It reads more like an invocation instruction than a description, failing to communicate what the skill does, what actions it performs, or when it should be used. It scores at the lowest level across all dimensions.
Suggestions
Add concrete actions describing what the skill does, e.g., 'Runs performance benchmarks on APIs/services, measures response times, throughput, and latency under load, and generates comparison reports.'
Add an explicit 'Use when...' clause with natural trigger terms, e.g., 'Use when the user asks about performance testing, benchmarking, load testing, measuring latency, throughput analysis, or stress testing.'
Remove the invocation instruction ('invoke with $agent-performance-benchmarker') from the description and replace it with capability and trigger information that helps Claude select the right skill.
| Dimension | Reasoning | Score |
|---|---|---|
Specificity | The description contains no concrete actions whatsoever. It only says 'Agent skill for performance-benchmarker' which is entirely vague and abstract, giving no indication of what the skill actually does. | 1 / 3 |
Completeness | The description fails to answer both 'what does this do' and 'when should Claude use it'. It provides neither capability information nor usage triggers — only an invocation command. | 1 / 3 |
Trigger Term Quality | The only potentially relevant term is 'performance-benchmarker', which is a tool name rather than a natural keyword a user would say. There are no natural trigger terms like 'benchmark', 'performance testing', 'load test', 'latency', etc. | 1 / 3 |
Distinctiveness Conflict Risk | The term 'performance-benchmarker' is too vague to carve out a clear niche. Without specifying what kind of performance (web, database, API, code) or what benchmarking entails, it could easily conflict with other performance-related skills. | 1 / 3 |
Total | 4 / 12 Passed |
Implementation
0%Reviews the quality of instructions and guidance provided to agents. Good implementation is clear, handles edge cases, and produces reliable results.
This skill is an extremely verbose, non-executable pseudocode dump that provides no actionable guidance for Claude. It explains concepts Claude already knows (statistical calculations, monitoring patterns) through hundreds of lines of fictional class implementations that reference non-existent dependencies. The content fails on every dimension: it wastes tokens, provides no executable code, lacks workflow structure, and has no progressive disclosure.
Suggestions
Replace the entire pseudocode implementation with a concise overview of benchmarking responsibilities and 2-3 short, actually executable code snippets or concrete command examples that Claude can use.
Define a clear step-by-step workflow: e.g., 1) Configure benchmark parameters, 2) Run benchmarks, 3) Validate results against thresholds, 4) Generate report — with explicit validation checkpoints.
Remove all references to fictional classes and MCP tools that don't exist, or replace them with real, usable tools and libraries.
Extract detailed reference material into separate bundle files and keep SKILL.md as a lean overview with navigation links to those files.
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | Extremely verbose at ~600+ lines of code that Claude doesn't need spelled out. The entire implementation is pseudocode-style class definitions for hypothetical classes (SystemMonitor, PerformanceModel, etc.) that don't actually exist. This explains concepts Claude already understands (how to calculate percentiles, standard deviations, etc.) and wastes enormous token budget on non-executable illustrative code. | 1 / 3 |
Actionability | Despite the massive amount of code, none of it is executable. It references non-existent classes (TimeSeriesDatabase, SystemMonitor, PerformanceModel, LoadGenerator, etc.) and fictional MCP tools (neural_patterns, neural_predict, metrics_collect). There are no real commands, no real libraries, and nothing copy-paste ready. This is elaborate pseudocode dressed up as implementation. | 1 / 3 |
Workflow Clarity | There is no clear workflow or sequence of steps for Claude to follow. The content is a collection of class definitions without any guidance on when to use them, in what order, or how to validate results. No validation checkpoints, no error recovery steps, no decision points are articulated as a workflow. | 1 / 3 |
Progressive Disclosure | The content is a monolithic wall of code with no structure for progressive disclosure. Everything is dumped into a single file with no references to supporting documents. The massive inline code blocks should be separated into reference files, with the SKILL.md providing a concise overview and navigation. | 1 / 3 |
Total | 4 / 12 Passed |
Validation
90%Checks the skill against the spec for correct structure and formatting. All validation checks must pass before discovery and implementation can be scored.
Validation — 10 / 11 Passed
Validation for skill structure
| Criteria | Description | Result |
|---|---|---|
skill_md_line_count | SKILL.md is long (856 lines); consider splitting into references/ and linking | Warning |
Total | 10 / 11 Passed | |
- Repository
- ruvnet/ruflo
- Commit
619b263
- Reviewed
Table of Contents
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.