Skill

rwsdk-audit-deployed

This skill should be used when the user asks to "audit my deployed app", "check production errors", "is my worker healthy", "check logs", "investigate production issues", "analyze worker performance", or any request to inspect a deployed RedwoodSDK/Cloudflare Worker application. Trigger on mentions of production errors, Cloudflare Worker logs, observability, metrics, latency issues, error rates, deployment health checks, or "pnpm release" follow-up verification. Also trigger when the user asks "is everything working" or wants to verify a recent deployment succeeded. Uses cloudflare-observability MCP for real-time worker telemetry and rwsdk-docs skill for best practice verification. Provides a structured 6-phase audit workflow with automated report generation.

Install

npx claudepluginhub softee-p/redwoodsdk-toolkit --plugin rwsdk

Tool Access

This skill uses the workspace's default tool permissions.

Preview

Methodically audit a deployed RedwoodSDK application on Cloudflare Workers.

Supporting Assets

references/common-issues.md

SKILL.md

Similar Skills

skill-lookup

Searches, retrieves, and installs Agent Skills from prompts.chat registry using MCP tools like search_skills and get_skill. Activates for finding skills, browsing catalogs, or extending Claude.

prompts.chat

159.9k

prompt-lookup

Searches prompts.chat for AI prompt templates by keyword or category, retrieves by ID with variable handling, and improves prompts via AI. Use for discovering or enhancing prompts.

prompts.chat

159.9k

MCP Integration

6 files

Guides MCP server integration in Claude Code plugins via .mcp.json or plugin.json configs for stdio, SSE, HTTP types, enabling external services as tools.

plugin-dev

83.2k

Stats

Parent Repo Stars0

Parent Repo Forks0

Last CommitMar 19, 2026

Actions

View Source View Plugin View on GitHub View README

RedwoodSDK Deployed Application Audit

Methodically audit a deployed RedwoodSDK application on Cloudflare Workers.

Prerequisites

cloudflare-observability MCP (required) - This skill depends on the cloudflare-observability MCP server for querying worker logs and metrics. The MCP must be configured in your Claude Code settings with valid Cloudflare API credentials. Without it, Phases 1-5 cannot execute.
rwsdk-docs skill - For verifying code against best practices

Audit Workflow

Execute these phases in order. Use TodoWrite to track progress.

Phase 1: Setup & Discovery

Auto-detect worker name (if in a project):
- Check wrangler.jsonc or wrangler.toml for name field
- If found, use it as default worker name
List workers: mcp__cloudflare-observability__workers_list
Confirm worker name with user if:
- Multiple workers exist
- No wrangler config found
- User specifies a different worker
Get worker details: mcp__cloudflare-observability__workers_get_worker

Phase 2: Error Analysis (24h)

Query recent errors:

{
  "view": "events",
  "queryId": "errors-24h",
  "limit": 20,
  "parameters": {
    "filters": [
      {"key": "$metadata.service", "operation": "eq", "type": "string", "value": "<worker-name>"},
      {"key": "$metadata.level", "operation": "eq", "type": "string", "value": "error"}
    ]
  },
  "timeframe": {"reference": "<current-iso-time>", "offset": "-24h"}
}

Analyze patterns:

Route-specific errors ($metadata.trigger)
Version-specific ($workers.scriptVersion.id) - different versions indicate deployment changes
Error fingerprints for grouping

Phase 3: Warning Analysis

Query warnings (cross-request promise issues common in rwsdk):

{
  "view": "events",
  "queryId": "warnings-24h",
  "limit": 10,
  "parameters": {
    "filters": [
      {"key": "$metadata.service", "operation": "eq", "type": "string", "value": "<worker-name>"},
      {"key": "$metadata.level", "operation": "eq", "type": "string", "value": "warn"}
    ]
  },
  "timeframe": {"reference": "<current-iso-time>", "offset": "-24h"}
}

Phase 4: Current Health Check

Verify no recent errors (confirms if issues resolved):

{
  "view": "calculations",
  "queryId": "errors-1h",
  "parameters": {
    "filters": [
      {"key": "$metadata.service", "operation": "eq", "type": "string", "value": "<worker-name>"},
      {"key": "$metadata.level", "operation": "eq", "type": "string", "value": "error"}
    ],
    "calculations": [{"operator": "count", "alias": "error_count"}]
  },
  "timeframe": {"reference": "<current-iso-time>", "offset": "-1h"}
}

Phase 5: Performance Metrics (7d)

Query health and performance by outcome:

{
  "view": "calculations",
  "queryId": "health-7d",
  "parameters": {
    "filters": [{"key": "$metadata.service", "operation": "eq", "type": "string", "value": "<worker-name>"}],
    "calculations": [
      {"operator": "count", "alias": "total"},
      {"operator": "avg", "key": "$workers.wallTimeMs", "keyType": "number", "alias": "avg_latency"},
      {"operator": "p99", "key": "$workers.wallTimeMs", "keyType": "number", "alias": "p99_latency"}
    ],
    "groupBys": [{"type": "string", "value": "$workers.outcome"}]
  },
  "timeframe": {"reference": "<current-iso-time>", "offset": "-7d"}
}

Phase 6: Code Quality (If Errors Found)

When errors reference code paths:

Use rwsdk-docs skill to verify patterns against docs
Read source files from error stack traces
Check common issues - See references/common-issues.md

Report Structure

# Audit Report: <worker-name>

## Executive Summary
**Status**: [HEALTHY | DEGRADED | CRITICAL]

## Health Metrics (7 Days)
| Outcome | Count | % |
|---------|-------|---|

**Performance:** Avg: Xms, P99: Xms

## Issues Found
### [RESOLVED | ACTIVE]: <Title>
**Error:** <message>
**Route:** <trigger>
**Root Cause:** <explanation>

## Performance by Route
| Route | Avg | P99 | Notes |

## Code Quality
| Pattern | Status | Notes |

## Recommendations

Alert Thresholds

Metric	Warning	Critical
Error rate	>2%	>5%
P99 latency	>3s	>5s
Canceled rate	>10%	>15%

Version mismatch in errors = Recent deployment likely fixed/caused issue