performance-monitor-analyst

将性能目标与实际的性能剖析结果进行关联分析，精准定位瓶颈并对照非功能性需求进行验证。

SKILL.md

--- frontmatter

name: performance-monitor-analyst
description: Correlates performance targets with actual profiling results. Identifies bottlenecks and validates against non-functional requirements.
status: implemented
arguments:
  - name: input
    short: i
    type: string
    required: true
    description: Path to a JSON file containing performance metrics
  - name: out
    short: o
    type: string
    description: Output file path

Performance Monitor Analyst

This skill compares "what we want" (NFR) with "what we have" (Profiling Logs).

Capabilities

1. Profiling Log Analysis

•Reads outputs from Clinic.js, cProfile, or chrome://tracing.
•Identifies heavy functions and memory leaks.

2. Gap Analysis

•Compares measured response times (Latency) and other Golden Signals against the targets in knowledge/nonfunctional/.
•Evaluates SLO attainment and remaining Error Budget based on SLO & Dashboard Best Practices.
•Issues "Warning" if targets are missed or if saturation levels are too high.

Usage

•"Analyze this profile.json and tell me if we are meeting our 200ms response time requirement."
•"Where is the bottleneck in this Python profile?"

Knowledge Protocol

•This skill adheres to the knowledge/orchestration/knowledge-protocol.md. It automatically integrates Public, Confidential (Company/Client), and Personal knowledge tiers, prioritizing the most specific secrets while ensuring no leaks to public outputs.
•References Monitoring Best Practices for standard performance metrics (Golden Signals).
•References SLO & Dashboard Best Practices for service level objectives and error budget analysis.