Twilio SMS Delivery Health Monitor

What it does

Twilio SMS Delivery Health Monitor collects SMS delivery statistics from the Twilio API and error-rate metrics from Datadog, analyzes delivery rates by carrier and error code, detects drops below configurable thresholds, generates deterministic root cause hypotheses, and dispatches corrective-action alerts to Slack with a 4-hour flood-prevention throttle. Every finding and hypothesis is code-derived; the LLM only narrates. Full GDPR guardrails: phone number masking and no message content logging. Built for platform and communications engineering teams who want automated SMS delivery health monitoring instead of manually cross-referencing Twilio and Datadog dashboards.

Example prompts

• Run the full SMS delivery health check and alert Slack if needed
• Show me the carrier-level delivery rate breakdown
• What error codes are causing SMS failures right now?
• Check if we can send another Slack alert or if it's throttled

Tools (8)

Tools the agent exposes — your AI client calls them automatically when it needs them.

• run_full — Run the full SMS delivery health check: collect Twilio + Datadog data, analyze delivery rate, error codes and carrier breakdown, synthesize LLM narrative summary.
• analyze_delivery — Given raw collected data, run pure deterministic analysis and return findings, root cause hypothesis and corrective actions. No LLM.
• get_carrier_breakdown — Collect SMS data and return carrier-level delivery rate breakdown.
• get_error_analysis — Collect SMS data and return a detailed error code analysis with recommended fixes.
• check_alert_status — Check the Slack alert throttle state.
• discover_intent — Map a natural-language user query to the most appropriate tool and pre-filled arguments.
• plan_inputs — Plan/brainstorm the inputs for a tool: returns questions, schema and a ready-to-edit example.
• list_capabilities — List the agent's capabilities: tools, connectors, credentials and guardrails.