Token导航 LogoToken导航TokenDH.com
研究检索需要联网github未标认证来源可访问许可证需确认审计通过

export-debugger导出调试器

Agent Skill

用于辅助前端页面、组件、样式和交互逻辑的开发与维护。它适合让 Agent 生成或审查 React、Next.js、Vue、Tailwind、CSS 等相关代码,整理组件结构,或定位布局和性能问题。使用时需要结合项目现有设计系统、路由和构建方式,避免只生成孤立片段;涉及页面改动时,应配合本地预览和构建检查确认视觉效果。

总安装

499

周安装

21

GitHub Stars

2

下载量

175
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:export-debugger(导出调试器)
来源仓库:https://github.com/lytics/agent-skills
仓库路径:skills/export-debugger
安装命令:
npx skills add https://github.com/lytics/agent-skills --skill export-debugger
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/lytics/agent-skills --skill export-debugger

简介

诊断用户数据是否成功导出至外部平台,追踪导出管道各环节状态。

  • 检查分段归属、任务状态、流程节点及静默窗口等关键信息。
  • 输入为用户标识(如邮箱)及导出目标(作业 ID 或平台名称)。
  • 仅提供只读分析,不涉及数据修改或导出操作本身执行。
  • export-debugger 属于研究检索类 Skill,可作为该场景下的辅助能力补充。

SKILL.md

Export Debugger

Purpose

Diagnoses why a specific user was or wasn't exported to an external platform. Traces through the full export pipeline: segment membership -> job status -> flow state -> quiet windows -> export logs. Read-only diagnostic.

Environment

  • LYTICS_API_TOKEN -- API authentication token
  • LYTICS_API_URL -- Base URL (default: https://api.lytics.io)

Inputs

  • User identity (field + value, e.g., email user@example.com)
  • Export destination: job ID, job name, platform name, or flow name
  • Table (default: user)

Diagnostic Flow

Step 1: Identify the Export Job

If the user provides a job ID, fetch it directly. Otherwise, find the relevant job:

# List all jobs to find the export
curl -s "${LYTICS_API_URL:-https://api.lytics.io}/v2/job" \
  -H "Authorization: ${LYTICS_API_TOKEN}"

Match by name, workflow type, or platform. Extract the segment_id from the job's config field -- this is the segment the job exports from.

If the user mentions a flow, fetch it and trace to the export step:

curl -s "${LYTICS_API_URL:-https://api.lytics.io}/v2/flow/ui/${FLOW_ID}" \
  -H "Authorization: ${LYTICS_API_TOKEN}"

Find steps with type_hint: "work_export" and get their work_id to identify the export job.

Step 2: Check User's Segment Membership

curl -s "${LYTICS_API_URL:-https://api.lytics.io}/api/entity/user/${FIELD}/${VALUE}?segments=true&allsegments=true" \
  -H "Authorization: ${LYTICS_API_TOKEN}"

Check: Is the user in the export job's target segment?

  • YES -> proceed to Step 3 (job should be exporting them)
  • NO -> this is likely the reason. Use profile-investigator skill to explain why they're not in the segment.

Step 3: Check Job Status

curl -s "${LYTICS_API_URL:-https://api.lytics.io}/v2/job/${JOB_ID}" \
  -H "Authorization: ${LYTICS_API_TOKEN}"
StatusDiagnosis
runnableJob is active -- check logs for export details
sleepingJob is between runs. Check sleep_until -- may be in quiet window or scheduled delay
pausedJob is paused. User or system paused it -- exports are halted
faultJob has errors. Check logs for the error details
failedJob has terminally failed. Check logs for root cause
killedJob was manually stopped

Step 4: Check Quiet Window

If the job has quiet window settings, calculate whether it's currently active:

From the job response, check these fields:

  • quiet_time_of_day -- e.g., "2:00pm"
  • quiet_timezone -- e.g., "America/Los_Angeles"
  • quiet_period -- hours of quiet, e.g., 4

If all three are set, calculate:

Quiet window: quiet_time_of_day to (quiet_time_of_day + quiet_period hours)
              in the configured timezone

Example: 2:00pm - 6:00pm America/Los_Angeles

If the current time falls within the quiet window, the job is sleeping and won't export until the window ends. Check sleep_until in the job response to confirm.

If drop_events_during_quiet_window is true, events arriving during quiet time are permanently dropped, not queued.

Step 5: Check Flow State (if flow-based export)

If the export is triggered through a flow, check which step the user is at:

The entity response from Step 2 includes flows_step_slugs:

{
  "flows_step_slugs": {
    "flow_id-1": "welcome_email_step"
  }
}

Cross-reference with the flow's step list to determine:

  • Has the user reached the export step yet? (may still be in a delay or earlier step)
  • Did the user exit the flow before reaching the export?
  • Is the user in a conditional branch that doesn't lead to the export?

Step 6: Check Job Logs

Always use include_errors=true when debugging to get full error details:

curl -s "${LYTICS_API_URL:-https://api.lytics.io}/v2/job/${JOB_ID}/logs?include_errors=true" \
  -H "Authorization: ${LYTICS_API_TOKEN}"

Look for:

  • Export counts: added, omitted, removed in the log context
  • Errors: Rate limits, auth failures, API errors from the destination platform
  • Timestamps: When the job last ran successfully
  • User-level errors: Some platforms report per-user failures (e.g., invalid email format)

Output Format

## Export Debug Report

### User: user@example.com
### Destination: Facebook Custom Audiences (job: "fb_export_high_value")

### Diagnostic Results

1. SEGMENT MEMBERSHIP: PASS
   User IS in segment "High Value Customers" (id: seg123)

2. JOB STATUS: PASS
   Job is "runnable" -- actively processing

3. QUIET WINDOW: FAIL
   Job is currently in quiet window (2:00pm - 6:00pm PT)
   Will resume at: 6:00pm PT (2026-03-19T01:00:00Z)
   drop_events_during_quiet_window: false (events are queued)

4. FLOW STATE: N/A
   Export is not flow-based

5. JOB LOGS: INFO
   Last export: 2026-03-19 09:45:00 UTC
   Added: 145, Omitted: 3, Removed: 12
   No errors in recent logs

### Diagnosis
The user IS in the export segment and the job is active, but the job
is currently in its quiet window (2:00pm-6:00pm PT). The user should
be exported when the job resumes at 6:00pm PT. Events are being
queued, not dropped.

Common Diagnoses

SymptomLikely CauseResolution
User not in segmentFilterQL conditions not metUse profile-investigator to see which conditions fail
Job pausedManually paused or auto-pausedResume the job
Job faultedAuth expired, rate limit, API errorCheck logs, fix auth or wait for rate limit reset
Job sleepingQuiet window or scheduled delayWait for quiet window to end
User in wrong flow stepDelay step or conditional branchCheck flow structure and timing
User exported but not appearingPlatform sync delayCheck platform-side, may take minutes to hours
"Omitted" in logsUser was filtered by export configCheck job config for additional filters beyond segment

Error Handling

  • Job not found: List jobs and help user identify the right one
  • User not found: Try alternative identity fields
  • Multiple export jobs for same platform: Show all, ask user to pick
  • No logs available: Job may be new or logs may have aged out

Dependencies

  • Composes: job-manager skill, entity-lookup skill, segment-manager skill, flow-manager skill
  • Related: profile-investigator skill (for segment membership diagnosis)
  • References: ../references/api-client.md

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

37.26%
按下载量换算65

Claude

28.26%
按下载量换算49

Cursor

17.15%
按下载量换算30

Gemini CLI

9.64%
按下载量换算17

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。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