agentleFS
mikeparcewskimikeparcewski/wicked-garden/skills/platform/errors/SKILL.md

wicked-garden-platform-errors

Skill9 starsChanged 7 months agoMITView on GitHub

Use when investigating a production error spike or pattern across services — aggregates errors from discovered tracking sources, correlates with deployments, and assesses user impact. NOT for general observability (use platform/observability) or tracing latency (use the platform:traces command).

The file

We split this skill into the rules it gives an agent

7 rules. About 0 are common, 6 are distinctive, and 1 is very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈6

    Distinctive · estimate

    This author's own choices. Worth a look.

  • 1

    Very specific

    Tied to their project's names, files or tools. Probably skip.

  • Common
  • Distinctive
  • Very specific
  1. Distinctive:

    Check for correlation with:

    line 61
  2. Very specific:

    Fallback: Search code for error patterns via the wicked-garden-search skill (catch blocks, error handling, throw statements).

    line 84
  3. Distinctive:

    When crew completes build phase:

    line 150
  4. Distinctive:

    When errors detected, run the engineering debugging workflow with context:

    line 163
  5. Distinctive:

    Always correlate with deployments and code changes

    line 182
  6. Distinctive:

    Use traces to understand error propagation

    line 184
  7. Distinctive:

    Consider user impact, not just error count

    line 185

≈ “Common” is an estimate: at least 1 in 5 of the 372,429 skill files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.

Keep what fits and drop the rest. Your version, ready for your agents.
Files
SKILL.md

Error Analysis Skill

Aggregate and analyze errors from discovered error tracking sources with pattern detection.

When to Use

  • Error spike investigation
  • Pattern detection across services
  • User impact assessment
  • Correlation with deployments
  • Error trend analysis
  • User asks "errors", "exceptions", "why failing"

Error Analysis Approach

1. Discover Error Tracking Sources

Use capability-based discovery:

# Discover available integrations via capability detection

# Scan for error tracking capabilities by analyzing server descriptions:
# - error-tracking capability: Dedicated exception/error tracking
# - apm capability: APM tools that include error tracking
# - logging capability: Log platforms with error search/aggregation

2. Aggregate Error Data

For each discovered source:

  • Current error rate vs baseline
  • Top errors by frequency
  • New vs recurring errors
  • User sessions affected
  • Error distribution by service/region

3. Detect Patterns

Look for:

  • Error spikes: Sudden increase in error rate
  • New errors: First seen in recent time window
  • Error clusters: Related errors happening together
  • User impact: Same users hitting multiple errors
  • Service correlation: Errors across dependent services

4. Correlate with Changes

Check for correlation with:

  • Recent deployments
  • Code changes via the wicked-garden-search skill
  • Infrastructure changes
  • Traffic patterns
  • External dependency changes

5. Provide Investigation Path

Based on patterns:

  • Root cause hypothesis
  • Investigation steps
  • Recommended actions (rollback, hotfix, etc.)
  • Integration with engineering for debugging

Integration Discovery

CapabilityWhat to Look ForProvides
error-trackingException tracking, crash reporting, error groupingStack traces, user context, grouping
apmPerformance monitoring with error tracking featuresErrors with performance context
loggingLog platforms with error filtering and searchError logs, patterns, search

Fallback: Search code for error patterns via the wicked-garden-search skill (catch blocks, error handling, throw statements).

Output Format

## Error Analysis Report

**Analysis Time**: {timestamp}
**Time Range**: {period analyzed}
**Data Sources**: {list of integrations}

### Error Summary

**Current Error Rate**: {rate} ({change} from baseline)
**Total Errors**: {count} in last {period}
**Unique Errors**: {count} distinct error types
**Affected Users**: {count or percentage}

### Top Errors

| Error | Count | Users | First Seen | Trend |
|-------|-------|-------|------------|-------|
| {message} | {count} | {users} | {time} | {↑↓→} |

### Pattern Detection

**Pattern: {Pattern Name}**
- Type: [ERROR_SPIKE | NEW_ERROR | CASCADING | USER_CLUSTER]
- Description: {what the pattern indicates}
- Affected: {services, users, regions}
- Started: {timestamp}
- Correlation: {deployment, code change, etc.}

### Investigation Path

**Hypothesis**: {most likely root cause}

**Evidence**:
1. {supporting evidence point}
2. {supporting evidence point}

**Next Steps**:
1. {specific action to take}
2. {specific action to take}

**Engage**: the engineering skill's debugging workflow (the `wicked-garden-engineering-debugging` module) for code-level root-cause analysis

Common Error Patterns

Four main patterns to detect. See refs/patterns.md for detailed analysis guides.

ERROR_SPIKE

Sudden increase in error rate (>2x baseline). Often deployment-related.

NEW_ERROR

Error never seen before. Indicates new code path or edge case.

CASCADING_FAILURE

One error causing downstream errors. Check service dependencies.

USER_CLUSTER

Same users experiencing multiple errors. User-specific data issue.

Integration with wicked-crew

When crew completes build phase:

  1. Compare error rates pre/post deployment
  2. Alert on new errors
  3. Detect error spikes
  4. Recommend rollback if critical

Emit events:

  • observe:error:spike:warning
  • observe:error:pattern:warning
  • observe:correlation:found:success

Integration with engineering

When errors detected, run the engineering debugging workflow with context:

  • Error stack traces from discovered sources
  • Log context from logging integrations
  • Recent code changes from git
  • Deployment timeline

Severity Classification

See refs/severity.md for detailed classification.

CRITICAL: Error rate >10x baseline, critical path errors, data corruption HIGH: Error rate >3x baseline, affecting >10% users, payment/security errors MEDIUM: Error rate >1.5x baseline, affecting <10% users, non-critical features LOW: Error rate <1.5x baseline, cosmetic issues, logging errors

Notes

  • Focus on error rate changes, not absolute numbers
  • New errors often more critical than recurring ones
  • Always correlate with deployments and code changes
  • Distinguish between symptoms and root causes
  • Use traces to understand error propagation
  • Consider user impact, not just error count

Download now (ZIP)

Does it work?

Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.

No reviews yet

People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.

  • It worked
  • Partly
  • It didn't
What did you use it for? e.g. Reviewing pull requests in a Rails app

Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.

Your trail

Files you open will collect here, so you can find them again.

Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.

Now do this for your own.

Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.