From dx-automation
Show operational dashboard for AI automation agents — DLQ depth, monthly token budget, and daily rate limit usage. Hub profile only — requires AWS credentials. Read-only.
npx claudepluginhub easingthemes/dx-aem-flow --plugin dx-automationThis skill uses the workspace's default tool permissions.
You display an operational dashboard showing current DLQ depth, token budget utilization, and rate limit usage. All commands are read-only — no mutations.
Monitors AI agent health across quality, cost, performance, and errors using Amplitude Agent Analytics queries. Delivers trends, recent failures, and actionable reports for instrumented projects.
Instruments Python AI agents (LangChain, LangGraph, CrewAI, LlamaIndex, Google ADK) with OpenTelemetry to send traces, logs, metrics to DataRobot for monitoring.
Queries Elastic traces, metrics, and logs to monitor LLM/agentic performance, token/costs, quality, and orchestration. For LLM monitoring, GenAI observability, AI cost/quality.
Share bugs, ideas, or general feedback.
You display an operational dashboard showing current DLQ depth, token budget utilization, and rate limit usage. All commands are read-only — no mutations.
Read .ai/automation/infra.json to get automationProfile, region, and table names.
Profile check: If automationProfile is consumer (or legacy pr-only/pr-delegation):
This repo uses the <profile> profile — DLQ, token budget, and rate limits are managed by the hub project.
Run /auto-status from the hub repo instead.
STOP.
Check AWS credentials are available:
aws sts get-caller-identity --query Account --output text 2>/dev/null || echo "NO_CREDENTIALS"
If no credentials: "AWS credentials not configured. Run aws configure or set AWS_ACCESS_KEY_ID + AWS_SECRET_ACCESS_KEY." STOP.
cd .ai/automation
node eval/process-dlq.js --depth
If DLQ depth > 0:
# Show message summaries (not full content)
node eval/process-dlq.js --list
cd .ai/automation
node eval/cost-report.js
cd .ai/automation
node eval/rate-limit-report.js
Present results as a combined dashboard:
## Automation Status
**DLQ:** <N> messages <(✓ empty | ⚠️ N messages pending)>
**Token budget:** <utilization>% of monthly cap (<tokens>/<cap>) — mode: <normal|suggest-only|halted>
**Rate limits today:** DoR <N>/20, PR Review <N>/50, PR Answer <N>/30
<If any DLQ messages:>
### DLQ Messages
<list of message summaries: timestamp, error type, function name>
Investigate: `cd .ai/automation && node eval/process-dlq.js`
<If budget > 80%:>
⚠️ Token budget at <N>% — approaching limit. Consider increasing `MONTHLY_TOKEN_CAP` pipeline variable.
<If budget halted:>
🚨 Token budget exhausted — all LLM calls blocked until next month.
/auto-status — Queries AWS for DLQ depth (0 messages), monthly token usage (45% of budget), and daily rate limit usage (120/500 calls). Reports all metrics as healthy with green indicators.
/auto-status (DLQ has messages) — Reports DLQ depth of 3 messages with metadata (timestamps, error types, function names). Does not print message contents for security. Suggests running node eval/process-dlq.js to investigate.
/auto-status (budget warning) — Reports token budget at 85% with 6 days remaining in the month. Warns: "Approaching limit — consider increasing MONTHLY_TOKEN_CAP pipeline variable." DLQ and rate limits are healthy.
"Access denied" when querying AWS Cause: AWS credentials are not configured or lack read permissions for SQS, CloudWatch, or DynamoDB. Fix: Configure AWS credentials and ensure the IAM user has read permissions for the automation resources. This skill is read-only — it never modifies anything.
Token budget shows 0% but agents are running
Cause: The rate limit table in DynamoDB may not be tracking usage correctly, or the MONTHLY_TOKEN_CAP value is not set.
Fix: Check the DYNAMODB_RATE_LIMIT_TABLE entries directly in DynamoDB. Verify that MONTHLY_TOKEN_CAP is set as a pipeline variable.
DLQ depth keeps growing
Cause: An agent is consistently failing (expired PAT, invalid API key, unhandled error in agent step).
Fix: Process DLQ messages to identify the pattern. Common causes: expired ADO PAT (run /auto-lambda-env to update), LLM API key rotation needed, or a bug in agent step code (fix and run /auto-deploy).