Perf Tools
by ckchzh
Trace Linux system performance bottlenecks in real time. Use when scanning disk I/O, monitoring latency, reporting CPU usage, alerting on memory leaks.
安装
claude skill add --url github.com/openclaw/skills/tree/main/skills/ckchzh/perf-tools文档
Perf Tools
Sysops toolkit v2.0.0 — scan, monitor, report, and manage system performance from the command line.
Commands
All commands follow the pattern: perf-tools <command> [input]
When called without input, each command displays its recent entries. When called with input, it records a new timestamped entry.
| Command | Description |
|---|---|
scan | Record or view scan entries |
monitor | Record or view monitor entries |
report | Record or view report entries |
alert | Record or view alert entries |
top | Record or view top process entries |
usage | Record or view usage entries |
check | Record or view check entries |
fix | Record or view fix entries |
cleanup | Record or view cleanup entries |
backup | Record or view backup entries |
restore | Record or view restore entries |
log | Record or view log entries |
benchmark | Record or view benchmark entries |
compare | Record or view compare entries |
stats | Summary statistics across all log files |
export <fmt> | Export all data (json, csv, or txt) |
search <term> | Search across all log entries |
recent | Show the 20 most recent activity log entries |
status | Health check — version, entry count, disk usage |
help | Show help with all available commands |
version | Print version string |
How It Works
Each domain command (scan, monitor, report, etc.) works in two modes:
- Read mode (no arguments): displays the last 20 entries from its log file
- Write mode (with arguments): appends a timestamped
YYYY-MM-DD HH:MM|<input>line to its log file and logs the action tohistory.log
The built-in utility commands (stats, export, search, recent, status) aggregate data across all log files for reporting and analysis.
Data Storage
All data is stored locally in ~/.local/share/perf-tools/:
- Each command writes to its own log file (e.g.,
scan.log,monitor.log,alert.log) history.logrecords all write operations with timestamps- Export files are saved as
export.json,export.csv, orexport.txt - No external network calls — everything stays on disk
Requirements
- Bash 4+ with
set -euo pipefail - Standard Unix utilities:
date,wc,du,grep,tail,sed,cat - No external dependencies or package installations needed
When to Use
- Scanning for bottlenecks — log disk I/O issues, CPU hotspots, or latency spikes with
scan - Real-time monitoring — track system metrics with
monitorand recordalertentries when thresholds are breached - Performance reporting — use
reportandbenchmarkto capture test results andcompareruns side by side - Operational maintenance — log
cleanup,backup, andrestoreactivities for compliance and audit trails - Troubleshooting workflows — record diagnostic checks with
checkand remediation actions withfix
Examples
# Record a performance scan
perf-tools scan "Disk I/O latency: 15ms avg on /dev/nvme0n1 (normal: 2ms)"
# Monitor system metrics
perf-tools monitor "Load avg: 4.2 3.8 3.1, CPU temp: 72°C"
# Log an alert
perf-tools alert "Memory leak detected: process java-app growing 50MB/hour"
# Record a benchmark result
perf-tools benchmark "fio randread: 45K IOPS, 180MB/s throughput"
# Compare before and after
perf-tools compare "Pre-tune: 12ms p99 latency → Post-tune: 4ms p99 latency"
# Log a cleanup operation
perf-tools cleanup "Removed 3.8GB of old core dumps from /var/crash"
# Export all data to CSV
perf-tools export csv
# Search for memory-related entries
perf-tools search memory
# View recent activity
perf-tools recent
# Check overall status
perf-tools status
Output
All commands return results to stdout. Redirect to a file if needed:
perf-tools stats > perf-summary.txt
perf-tools export json
Powered by BytesAgain | bytesagain.com | hello@bytesagain.com
相关 Skills
安全专家
by alirezarezvani
覆盖威胁建模、漏洞评估、安全架构设计、代码审计与渗透测试,内置 STRIDE、OWASP、加密模式和安全扫描流程,适合系统设计评审与上线前安全排查。
✎ 安全专家把威胁建模、漏洞分析到渗透测试串成一套流程,内置 STRIDE 与 OWASP 指南,做安全设计和排查更省心。
安全运营
by alirezarezvani
覆盖应用安全、漏洞管理与合规审计,支持代码/依赖扫描、CVE 评估、Secrets 检测和安全自动化,适合做安全基线落地、漏洞响应、审计检查与安全开发治理。
✎ 应用安全、漏洞管理和合规检查一套打通,还能自动化扫描与响应,帮团队更早发现并收敛风险。
安全审计
by alirezarezvani
安装前审计 Claude Code Skill 的代码执行、Prompt 注入和依赖供应链风险,支持本地目录或 Git 仓库扫描,输出 PASS/WARN/FAIL 结论及修复建议
✎ 把代码审查、漏洞扫描和合规检查串成一条线,帮团队更早发现风险,做安全治理更省心。
相关 MCP 服务
by Sentry
搜索和分析 Sentry 错误报告,辅助调试。
✎ 把零散的 Sentry 错误报告变成可检索线索,帮你在海量报错里更快定位线上故障,排障调试明显省时。
by sinewaveai
为 AI agents 提供安全层:拦截 prompt injection、识别伪造 packages,并扫描漏洞风险。
✎ 给 AI Agent 补上关键安全层,能拦截 prompt 注入、识别伪造包并扫描漏洞风险,把防护前置更省心。
by pantheon-security
强化安全性的 NotebookLM MCP,集成 post-quantum encryption,提升数据防护能力。