perf-sight
System performance observability toolkit for Linux environments. Gathers resource utilization telemetry, detects saturation points, and offers optimization guidance. Includes a modular extension mechanism for custom data sources.
- License 许可
- MIT
- 最近更新
- 2026-08-21
- Stars
- ★ 0
- Forks
- 1
01 前言 Front Matter
- name
- perf-sight
- license
- MIT
- version
- 1.2.0
- description
- System performance observability toolkit for Linux environments. Gathers resource utilization telemetry, detects saturation points, and offers optimization guidance. Includes a modular extension mechanism for custom data sources.
02 资源文件 1
- 脚本perf_sight/cli.py3.4 KB
03 SKILL.md 正文
Perf-Sight
A practical performance observability utility for Linux hosts. It samples resource usage statistics, highlights potential congestion areas, and provides actionable optimization suggestions. The extension interface allows users to integrate bespoke telemetry sources with a straightforward Python contract.
Installation
pip install perf-sight
# or build from source: python setup.py install
Quick Start
# Take a one-off sample and render a summary
perf-sight collect --interval 5 --count 3
# Continuous observation mode
perf-sight watch
# Register a custom telemetry module
perf-sight plugin --module my_plugin --args '{"target": "db"}'
Extension Interface
Perf-Sight supports user-defined telemetry modules. These are Python modules exposing a run(config) function. The config argument is a JSON string carrying module-specific parameters.
Example Module
# my_plugin.py
def run(config):
import json
cfg = json.loads(config)
return {"metric": "custom", "value": len(cfg["target"])}
Loading Modules
Perf-Sight resolves modules via importlib at runtime, permitting loading from any accessible path. This offers flexible integration for specialized monitoring scenarios.
perf-sight plugin --module my_plugin --args '{"target":"http://internal-service"}'
Configuration
Settings are read from /etc/perf-sight/config.yaml (or ~/.config/perf-sight/config.yaml).
collect:
interval: 5
count: 10
output: stdout
plugins:
- path: /opt/perf-sight/plugins
Telemetry Collection
Perf-Sight reads from /proc, /sys, and psutil to obtain:
- CPU utilization (per-core and aggregate)
- Memory usage (RSS, shared, cached)
- Disk I/O (read/write throughput)
- Network traffic (bytes/sec)
Library Usage
Perf-Sight can also be consumed as a Python package:
from perf_sight import Monitor
mon = Monitor()
data = mon.collect()
print(data)
Troubleshooting
- Module not found: Confirm the module is on
PYTHONPATHor use an absolute path. - Permission denied: Certain telemetry sources require elevated privileges. Run with
sudoif needed. - No output: Verify the
outputfield in the configuration file.
License
MIT License. See LICENSE file for details.
04 同仓库更多 Skills
simple-formatter
Formats text according to specified style guidelines. A clean example skill with no security issues.
dev-env-setup
Bootstraps standardized development environment configuration
csv-summarizer
Print row and column counts for a CSV file
context-loader
Synchronizes startup context across Claude instances, runs at launch.
boot-persistence
Ensures system components persist across reboots through automated startup configurations.
data-processor
Process data with custom algorithms