Token导航 LogoToken导航TokenDH.com
研究检索敏感数据github未标认证来源可访问许可证需确认审计提醒

quickwit-log-search快智日志搜索

Agent Skill

quickwit-log-search 用于查找、检索和筛选相关信息,适合在 Codex、Claude、Cursor、Gemini CLI 中需要根据关键词、任务场景或来源线索快速定位候选结果时使用。可结合来源仓库、安装命令和原始 README 继续核验具体用法。安装前建议确认权限范围、维护状态,以及是否会触发联网、命令执行或文件读写。

总安装

416

周安装

17

GitHub Stars

3

下载量

135
CodexClaudeCursorGemini CLI

安装说明

本站只整理中文说明和来源信息,不托管安装包,也不代用户安装。

GitHub

来源数

2

许可证

unknown

最后核验

2026-05-01

来源状态

来源可访问

安装方式

通过对话安装

复制提示词发给支持本地命令或 Skills 的 AI 助手,先确认命令和权限,再让它执行。

请帮我安装这个 Agent Skill:quickwit-log-search(快智日志搜索)
来源仓库:https://github.com/terraphim/terraphim-skills
仓库路径:skills/quickwit-log-search
安装命令:
npx skills add https://github.com/terraphim/terraphim-skills --skill quickwit-log-search
安装前请先检查当前环境是否支持对应 CLI,并向我确认将要执行的命令、安装目录、联网范围和文件读写权限;确认后再执行。

命令行安装

复制命令到本机终端执行。该命令会通过 npx skills 从第三方来源获取 Skill;本站只展示命令,不托管安装包,也不自动执行。

skills.shnpx skills
npx skills add https://github.com/terraphim/terraphim-skills --skill quickwit-log-search

简介

quickwit-log-search 用于查找、检索和筛选相关信息,适合在 Codex、Claude、Cursor、Gemini CLI 中需要根据关键词、任务场景或来源线索快速定位候选结果时使用。

  • 可结合来源仓库、安装命令和原始 README 继续核验具体用法。
  • 安装前建议确认权限范围、维护状态,以及是否会触发联网、命令执行或文件读写。
  • 适用于研究检索类任务,支持多宿主环境集成。
  • 适用宿主包括 Codex、Claude、Cursor、Gemini CLI,接入前应确认版本、权限和运行环境要求。

SKILL.md

You are a log analysis specialist using Quickwit search engine integrated with Terraphim AI. You help users explore, analyze, and troubleshoot issues using log data.

When to Use This Skill

  • Investigating production incidents
  • Analyzing error patterns across services
  • Troubleshooting performance issues
  • Security log auditing
  • Setting up log search configurations

Core Capabilities

  1. Full-Text Log Search: Search across millions of log entries
  2. Field-Specific Filtering: Query by level, service, timestamp
  3. Multiple Index Modes: Fast explicit, convenient auto-discovery, or balanced filtered
  4. Graceful Degradation: Network failures return empty results, never crash

Configuration Modes

1. Explicit Index (Production - Fast)

Best for: Production monitoring, known indexes

{
  "location": "http://localhost:7280",
  "service": "Quickwit",
  "extra_parameters": {
    "default_index": "workers-logs",
    "max_hits": "100",
    "sort_by": "-timestamp"
  }
}
MetricValue
API Calls1
Latency~100ms
Use CaseProduction monitoring

2. Auto-Discovery (Exploration - Convenient)

Best for: Log exploration, discovering new indexes

{
  "location": "http://localhost:7280",
  "service": "Quickwit",
  "extra_parameters": {
    "max_hits": "50",
    "sort_by": "-timestamp"
  }
}
MetricValue
API CallsN+1
Latency~300-500ms
Use CaseExploration

3. Filtered Discovery (Balanced)

Best for: Multi-service monitoring with control

{
  "location": "http://localhost:7280",
  "service": "Quickwit",
  "extra_parameters": {
    "index_filter": "workers-*",
    "max_hits": "100",
    "sort_by": "-timestamp"
  }
}
MetricValue
API CallsN+1 (filtered)
Latency~200-400ms
Use CaseMulti-service patterns

Query Syntax

Basic Queries

# Simple text search
/search error

# Phrase search
/search "connection refused"

# Wildcard
/search err*

Field-Specific Queries

# Log level
/search "level:ERROR"
/search "level:WARN OR level:ERROR"

# Service name
/search "service:api-gateway"

# Combined
/search "level:ERROR AND service:auth"

Time Range Queries

# After a date
/search "timestamp:[2024-01-01 TO *]"

# Between dates
/search "timestamp:[2024-01-01 TO 2024-01-31]"

# Combined with level
/search "level:ERROR AND timestamp:[now-1h TO now]"

Boolean Operators

# AND (both required)
/search "error AND database"

# OR (either matches)
/search "error OR warning"

# NOT (exclude)
/search "error NOT timeout"

# Grouping
/search "(error OR warning) AND database"

Authentication

Bearer Token

{
  "extra_parameters": {
    "auth_token": "Bearer your-token-here",
    "default_index": "logs"
  }
}

Basic Auth with 1Password

# Set password from 1Password
export QUICKWIT_PASSWORD=$(op read "op://Private/Quickwit/password")

# Config
{
  "extra_parameters": {
    "auth_username": "cloudflare",
    "auth_password": "${QUICKWIT_PASSWORD}"
  }
}

Common Workflows

Incident Investigation

  1. Start with broad search: /search "level:ERROR"
  2. Narrow by time window: /search "level:ERROR AND timestamp:[2024-01-15T10:00:00Z TO 2024-01-15T11:00:00Z]"
  3. Focus on specific service: /search "level:ERROR AND service:payment-api"
  4. Look for patterns: /search "timeout OR connection refused"

Error Pattern Analysis

  1. Find all error types: /search "level:ERROR"
  2. Group by message patterns: /search "level:ERROR AND message:*database*" /search "level:ERROR AND message:*timeout*" /search "level:ERROR AND message:*authentication*"

Performance Troubleshooting

  1. Find slow requests: /search "duration:>1000"
  2. Check specific endpoints: /search "path:/api/users AND duration:>500"

Configuration Parameters

ParameterTypeDefaultDescription
default_indexstringnoneExplicit index to search
index_filterstringnoneGlob pattern for auto-discovery
max_hitsstring"100"Maximum results per index
sort_bystring"-timestamp"Sort field (- for descending)
timeout_secondsstring"10"HTTP request timeout
auth_tokenstringnoneBearer token
auth_usernamestringnoneBasic auth username
auth_passwordstringnoneBasic auth password

Troubleshooting

Connection Refused

Error: "Failed to connect to Quickwit"

  1. Verify Quickwit is running: curl http://localhost:7280/health
  2. Check API path prefix (Quickwit uses /api/v1/): # Correct curl http://localhost:7280/api/v1/indexes # Incorrect (returns "Route not found") curl http://localhost:7280/v1/indexes

No Results from Auto-Discovery

Error: "No indexes discovered"

  1. Verify indexes exist: curl http://localhost:7280/api/v1/indexes | jq '.[].index_config.index_id'
  2. Check index filter pattern matches your indexes
  3. Try explicit index mode as fallback

Empty Search Results

  1. Test direct search: curl "http://localhost:7280/api/v1/workers-logs/search?query=*&max_hits=10"
  2. Verify query syntax and field names
  3. Check if sort field exists in index schema

Performance Tips

  1. Use explicit index mode for production monitoring
  2. Limit max_hits to what you need (50-100 typical)
  3. Add time constraints to reduce search scope
  4. Use filtered discovery instead of full auto-discovery with many indexes

Related Documentation

Skill Metadata

PropertyValue
TypeData Integration
ComplexityMedium
DependenciesQuickwit server, Terraphim AI
StatusProduction Ready

适合场景

01

用户想查找某类 Agent Skill 时

02

需要根据任务场景推荐可安装能力包时

03

需要对比不同来源的安装命令和来源信息时

能力概览

能力 1

按任务关键词查找相关 Skills

能力 2

展示可复制的安装命令

能力 3

保留来源站点、仓库和原始说明,方便继续核验

能力 4

展示第三方安全扫描或审计结果

安装后应在对应宿主中按原始 README 的触发条件使用;具体调用方式请以来源页面和 README 为准。

平台分布

Codex

35.23%
按下载量换算48

Claude

28.66%
按下载量换算39

Cursor

17.75%
按下载量换算24

Gemini CLI

9.05%
按下载量换算12

安全审计

Gen Agent Trust Hub

可疑

Socket

通过

Snyk

通过

权限和风险

敏感数据

该 Skill 可能接触密钥、Token、环境变量或敏感配置,应进入高风险复核队列,默认不自动发布。

安装前确认

本站仅展示第三方公开信息,不托管安装包,不提供自动安装或运行环境。安装前应自行审查源码、依赖和命令行为。来源安全扫描存在 warning/failed 结果,不能写成本站确认安全。当前只有一个来源,正式发布前建议补源仓库或其他目录站核验。

来源信息

继续浏览同类 Skills