Token导航 LogoToken导航

Agent Skill · 研究检索

log-reader(日志阅读器)

log-reader 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。

log-reader 是研究检索类 Agent Skill,可用宿主为 Codex、Claude、Cursor、Gemini CLI,通过GitHub 仓库获取,权限为需要联网;许可证为 MIT,第三方扫描已通过,按安装量在研究检索类的 73,675 个 Skill 里排第 35,605。

总安装 2,056周安装 84GitHub Stars 38下载 665核验 2026-05-01

安装说明

本站只整理中文说明,不托管安装包,也不代用户安装。
GitHub
许可证
MIT

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:log-reader(日志阅读器)
来源仓库:https://github.com/terrylica/cc-skills
仓库路径:skills/log-reader
安装命令:
npx skills add https://github.com/terrylica/cc-skills --skill log-reader
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。不同来源提供的安装方式可能略有差异;本站展示可直接复制的安装命令,安装前请核对来源页面。

skills.shnpx skills
npx skills add https://github.com/terrylica/cc-skills --skill log-reader

这个 Skill 是做什么的

研究检索类 Skill 主要解决「找资料、核对事实、把来源整理清楚」的问题,通常会让助手先检索再总结。

用于查找、检索和筛选相关信息,支持关键词和任务场景匹配。适合快速定位候选结果,提升信息获取效率。log-reader 属于研究检索类 Skill,可作为该场景下的辅助能力补充。可通过来源仓库和 README 进一步验证具体用法。安装前建议确认权限范围、维护状态及是否触发联网或文件读写。

适用于 Codex、Claude、Cursor 和 Gemini CLI 等宿主环境。

来源仓库内路径:skills/log-reader

上手步骤

4 步
  1. 1

    确认自己的宿主工具

    公开资料标注支持 Codex、Claude、Cursor、Gemini CLI;先确认自己用的是其中之一,再按对应工具的 Skills 目录放置。

  2. 2

    拿到安装命令

    可以直接复制的命令是 npx skills add https://github.com/terrylica/cc-skills --skill log-reader;命令里的占位符换成自己的路径或密钥再执行。

  3. 3

    安装后重启宿主

    命令执行完要完全退出宿主工具再打开,让新装的 Skill 被重新扫描到,否则列表里可能看不到。

  4. 4

    按最小权限试用

    它需要需要联网,先在测试目录或测试账号里跑,确认没问题再放到正式项目。

常见问题

6 条

log-reader 怎么安装?

在终端执行 npx skills add https://github.com/terrylica/cc-skills --skill log-reader。装完完全退出宿主工具再打开,让它重新扫描本地 Skill。

log-reader 支持哪些 AI 工具?

公开资料标注支持 Codex、Claude、Cursor、Gemini CLI。其他工具只要能读取本地 Skill 目录,一般也能手动放进去,但要自己确认触发方式是否一致。

log-reader 需要什么权限?

权限级别是需要联网。该 Skill 可能需要联网访问来源站点、仓库或外部 API;具体网络访问范围需要结合源码和 README 复核。

log-reader 是免费的吗?

来源标注的许可证是 MIT。本站不代收任何费用,也不托管安装包;能否商用、要不要署名,以许可证原文为准。

log-reader 用的人多吗?

公开数据是累计安装 2,056 次,近一周安装 84 次,下载 665 次,GitHub 38 星。这些数字来自第三方目录统计,只能当作热度参考,不代表质量或安全性。

log-reader 安全吗?

第三方扫描结果为通过,但这只覆盖已知的静态检查项,不能替代自己读一遍源码和安装命令。

值得选的地方和要先想清楚的地方

值得选的地方

  • 同时适配 4 种工具,换客户端不用重新找替代品。
  • 3 个目录同时收录,说明这个 Skill 流传比较广。
  • 许可证明确为 MIT,商用和分发边界好判断。
  • 第三方扫描已通过,常见的高风险行为没有命中。

先想清楚的地方

  • 第三方目录整理,版本更新和行为细节以来源仓库为准。

适不适合你

更适合

  • 已经在用 Codex,想直接扩展它能力的人。
  • 需要处理研究检索类任务、又不想每次都手写提示词的人。
  • 需要许可证清楚、并且有第三方扫描结果的团队项目。

不太适合

  • 需要厂商级技术支持和 SLA 的生产系统。

研究检索类 Skill 对比

6 个同类
Skill总安装GitHub 星权限许可证扫描
find-skills

寻找技能 · Codex、Claude

31,199,81316,486操作浏览器MIT有提醒
self-improving-agent

自我改进 Agent · OpenClaw

16,464,0023,396接触敏感数据MIT-0通过
soultrace

灵魂追踪 · Codex、Claude

7,588,94412需要联网未明确有提醒
self-improving

自我完善 · OpenClaw

6,829,2841,046需要联网MIT-0通过
microsoft-foundry

微软代工厂 · Codex、Claude

6,614,144764调用外部服务未明确有提醒
log-reader

日志阅读器 · 本页

2,05638需要联网MIT通过

按安装量取同分类前几名;数字来自第三方目录统计,只作热度参考,不代表质量结论。

SKILL.md 原文

MT5 Log Reader

Read MetaTrader 5 log files directly to access Print() output from indicators, scripts, and expert advisors without requiring manual Experts pane inspection.

Self-Evolving Skill: This skill improves through use. If instructions are wrong, parameters drifted, or a workaround was needed — fix this file immediately, don't defer. Only update for real, reproducible issues.

Purpose

Implement "Option 3" dual logging pattern:

  • Print() - MT5 log files (human-readable via Experts pane)
  • CSV files - Structured data (programmatic analysis)

Claude Code can autonomously read both outputs without user intervention.

When to Use This Skill

Use this skill when:

  • Validating MT5 indicator/script execution
  • Checking compilation or runtime errors
  • Analyzing Print() debug output
  • Verifying unit test results (Test_PatternDetector, Test_ArrowManager)
  • User mentions checking "Experts pane" manually

Log File Location

MT5 logs are stored at:

$MQL5_ROOT/Program Files/MetaTrader 5/MQL5/Logs/YYYYMMDD.log

File Format:

  • Encoding: UTF-16LE (Little Endian)
  • Structure: Tab-separated fields (timestamp, source, message)
  • Size: Grows throughout day (typically 10-100KB)

Instructions

1. Construct today's log path

/usr/bin/env bash << 'SKILL_SCRIPT_EOF'
# Determine current date
TODAY=$(date +"%Y%m%d")

# Build absolute path
LOG_FILE="$MQL5_ROOT/Program Files/MetaTrader 5/MQL5/Logs/${TODAY}.log"
SKILL_SCRIPT_EOF

2. Read the entire log file

Use Read tool:

  • File path: Absolute path from step 1
  • The file contains all Print() statements from MT5 indicators/scripts
  • UTF-16LE encoding is automatically handled by Read tool

3. Search for specific content (optional)

Use Grep to filter entries:

Pattern: indicator name, "error", "test.*passed", etc.
Path: Log file path from step 1
Output mode: "content" with -n (line numbers)
Context: -A 5 for 5 lines after matches

4. Analyze recent entries (optional)

Use Bash with tail for latest output:

tail -n 50 "$LOG_FILE"

Common Validation Patterns

Check unit test results

Search for test pass/fail indicators:

Pattern: test.*passed|test.*failed|Tests Passed|Tests Failed|ALL TESTS PASSED
Output mode: content
Context: -B 2 -A 2

Find compilation errors

Pattern: error|ERROR|warning|WARNING|failed to create
Output mode: content
Context: -A 3

Monitor specific indicator

Pattern: CCI Rising Test|PatternDetector|ArrowManager
Output mode: content
Context: -A 2

View initialization messages

Pattern: OnInit|initialization|Initialization complete|Phase \d+
Output mode: content

Examples

Example 1: Validate unit test completion

Input: User compiled Test_PatternDetector.mq5
Action:
  1. Read today's log file
  2. Grep for "Test.*PatternDetector|Tests Passed|Tests Failed"
  3. Report results (e.g., "17 tests passed, 0 failed")
Output: Test status without user checking Experts pane

Example 2: Check for runtime errors

Input: User reports indicator not working
Action:
  1. Read today's log file
  2. Grep for "ERROR|error|failed" with -A 3 context
  3. Analyze error messages
Output: Specific error details and line numbers

Example 3: Verify Phase 2 arrow creation

Input: User asks "did the test arrow get created?"
Action:
  1. Read today's log file
  2. Grep for "Phase 2|Test arrow created|Failed to create"
  3. Check for success/failure messages
Output: Arrow creation status with timestamp

Security Considerations

  • Log files may contain sensitive trading data (symbol names, account info)
  • Restricted to Read, Bash, Grep tools only (no network access via WebFetch)
  • Do not expose absolute paths unnecessarily in user-facing output
  • Filter sensitive information when reporting results
  • No file modification operations allowed

Integration with Dual Logging

This skill enables programmatic access to one half of the dual logging pattern:

  1. MT5 Log Files (this skill) - Human-readable Print() output
  2. CSV Files (CSVLogger.mqh) - Structured audit trails for validation

Both are accessible without user intervention:

  • MT5 logs: Read via this skill
  • CSV files: Read directly via Read tool or validate_export.py

Validation Checklist

When using this skill:

  • Log file exists for today's date
  • File size > 0 (not empty)
  • Contains expected indicator/script output
  • Timestamps match execution time
  • Error messages (if any) are actionable
  • Test results (if applicable) show pass/fail counts

References

  • MT5 file locations: docs/guides/MT5_FILE_LOCATIONS.md
  • Dual logging implementation: docs/plans/cci-rising-pattern-marker.yaml Phase 3-4
  • CSVLogger library: Program Files/MetaTrader 5/MQL5/Indicators/Custom/Development/CCINeutrality/lib/CSVLogger.mqh

Troubleshooting

IssueCauseSolution
Log file not foundWrong date or pathVerify YYYYMMDD.log format and MQL5_ROOT env var
Empty log fileMT5 not running or no outputEnsure MT5 is running and Print() is being called
Encoding errorsUTF-16LE not handledRead tool handles encoding automatically
Missing test resultsTest not executedCompile and run test script in MT5 first
Grep finds nothingWrong patternUse case-insensitive (-i) or broader pattern
Old log dataLog rotationEach day creates new YYYYMMDD.log file
Path contains spacesUnquoted path variableQuote paths: "$LOG_FILE"
Sensitive data exposedTrading info in logsFilter sensitive fields when reporting to user

Post-Execution Reflection

After this skill completes, check before closing:

  1. Did the command succeed? — If not, fix the instruction or error table that caused the failure.
  2. Did parameters or output change? — If the underlying tool's interface drifted, update Usage examples and Parameters table to match.
  3. Was a workaround needed? — If you had to improvise (different flags, extra steps), update this SKILL.md so the next invocation doesn't need the same workaround.

Only update if the issue is real and reproducible — not speculative.

平台分布

Claude Code

28.84%
按下载量换算192

OpenCode

23.35%
按下载量换算155

Antigravity

17.66%
按下载量换算117

Gemini CLI

12.27%
按下载量换算82

windsurf

8.58%
按下载量换算57

trae

3.68%
按下载量换算24

安全审计

Gen Agent Trust Hub

通过

Socket

通过

Snyk

通过

权限和风险

需要联网

该 Skill 可能需要联网访问来源站点、仓库或外部 API;具体网络访问范围需要结合源码和 README 复核。

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。

继续浏览同类 Skills