AgentSkillsCN

Cloudflare Workers Observability

当用户询问“工作者日志”、“调试工作者”、“工作者错误”、“请求分析”、“工作者指标”、“性能监控”、“错误率”、“调用日志”、“排查工作者”、“工作者分析”时，或需要调试和监控Cloudflare Workers时，应使用此技能。

SKILL.md
--- frontmatter
name: Cloudflare Workers Observability
description: This skill should be used when the user asks about "worker logs", "debug worker", "worker errors", "request analytics", "worker metrics", "performance monitoring", "error rate", "invocation logs", "troubleshoot worker", "worker analytics", or needs to debug and monitor Cloudflare Workers.
version: 1.0.0

Cloudflare Workers Observability

Debug and monitor Cloudflare Workers using logs and analytics from the Observability MCP server.

Available Tools

ToolPurpose
query_worker_observabilityQuery logs and metrics from Workers
observability_keysDiscover available data fields in logs
observability_valuesFind available values for specific fields

Query Workflow

1. Discover Available Fields

Use observability_keys to find what data is available:

  • •Metadata fields (timestamps, status codes)
  • •Worker-specific fields (script name, route)
  • •Custom logged fields from console.log

2. Explore Field Values

Use observability_values to find valid values for filtering:

  • •Status codes present in logs
  • •Script names deployed
  • •Custom field values

3. Query Logs and Metrics

Use query_worker_observability to:

  • •List recent events/invocations
  • •Calculate metrics (error rates, latency)
  • •Find specific invocations by criteria

Common Queries

GoalApproach
Recent errorsQuery for events with error status
Latency analysisQuery for execution time metrics
Traffic patternsQuery for invocation counts over time
Specific requestQuery by request ID or timestamp
Script comparisonQuery metrics grouped by script name

Debugging Workflow

  1. •

    Identify the problem

    • •Query recent errors with query_worker_observability
  2. •

    Find patterns

    • •Use observability_keys to discover relevant fields
    • •Use observability_values to see error types
  3. •

    Narrow down

    • •Add filters for specific routes, times, or status codes
  4. •

    Analyze specific invocations

    • •Query for detailed logs of problematic requests

Post-Deployment Monitoring

After deploying, check for issues:

  1. •Query for errors in the last 5-10 minutes
  2. •Compare error rates before/after deployment
  3. •Check latency metrics for performance regression

Tips

  • •Start broad, then add filters to narrow results
  • •Use observability_keys when unsure what fields exist
  • •Custom console.log output appears in queryable fields
  • •Combine with builds tools to correlate issues with deployments