OpenSkillIndex← back to search
claude-code

Forgent performance-bottleneck-detection

Performance bottleneck identification through code pattern analysis-based skill consuming source_code to produce performance_report

install
source · Clone the upstream repo
git clone https://github.com/mirandaguillaume/forgent
Claude Code · Install into ~/.claude/skills/
T=$(mktemp -d) && git clone --depth=1 https://github.com/mirandaguillaume/forgent "$T" && mkdir -p ~/.claude/skills && cp -r "$T/internal/bench/fixtures/imported/wshobson/output-standard/skills/performance-bottleneck-detection" ~/.claude/skills/mirandaguillaume-forgent-performance-bottleneck-detection && rm -rf "$T"
manifest: internal/bench/fixtures/imported/wshobson/output-standard/skills/performance-bottleneck-detection/SKILL.md
source content

Performance Bottleneck Detection

Guardrails

  • timeout: 180s
  • max_query_analysis_time: 60s
  • performance_threshold_ms: 100

Context

Consumes: source_code Produces: performance_report Memory: short-term

Strategy

Approach: Performance bottleneck identification through code pattern analysis Tools: read_file, grep, search, bash

Steps

  1. Analyze database queries for N+1 problems
  2. Review memory usage patterns
  3. Check caching implementations
  4. Assess async/concurrent patterns
  5. Evaluate resource management
  6. Review load handling strategies

Security

  • Filesystem: read-only
  • Network: none