# Diagnose Messaging Performance

Delivery data is difficult to interpret when statuses, channels, and webhook evidence disagree. This skill builds reliable cohorts and identifies the strongest supported cause.

## Install

```bash
npx skillstore add sentdm/messaging-performance-analyzer
```

## Metadata

- Status: approved
- Slug: sentdm-messaging-performance-analyzer
- Skillstore revision: r1
- Version status: missing
- Tree hash: 884961cb890fe9a813ac5889b1cd60ab7249aee2c8d328c6677cc134a93224db
- Author: sentdm
- GitHub username: sentdm
- License: MIT
- Repository: https://github.com/sentdm/sent-plugin/tree/bfc21e75db359a9c4022b4654055b04830342cbc/skills/messaging-performance-analyzer
- Ref: 403a72913d563cb23b8c54ebfae13b3ab6ebdb34
- Supported tools: Claude, Codex, Claude Code
- Audit status: complete
- Agent install advisory: allowed
- Manual install advisory: allowed
- Artifact signature: available
- Audit attestation: unavailable
- Human verification: not\_verified
- Risk factors: external\_commands
- Quality score: 80
- Quality tier: silver
- Public page: https://skillstore.pages.dev/skills/sentdm-messaging-performance-analyzer
- Manifest: https://skillstore.pages.dev/api/skills/sentdm-messaging-performance-analyzer/manifest

## Capabilities

- Analyzes Sent message records from CSV or JSON exports.
- Calculates observed lifecycle transition rates without inventing missing stages.
- Separates delivery outcomes by channel and direction.
- Reports WhatsApp and RCS read rates separately from delivery rates.
- Maps documented Sent and provider errors to focused verification steps.

## Use Cases

- Investigate delivery drops: Compare channel funnels, isolate the largest transition loss, and identify the evidence needed for a reliable diagnosis.
- Debug missing webhook events: Distinguish message delivery problems from webhook configuration, subscription, or endpoint failures.
- Review campaign performance: Measure delivery and engagement by profile, template, country, channel, and time window.

## Prompt Templates

### Summarize a delivery export

```
Analyze this Sent MDR export. Report counts by channel and latest status, then identify the largest supported delivery issue.
```

### Compare campaign periods

```
Compare these two campaign periods by channel, profile, template, and country. Show count and rate changes, then state the strongest supported explanation.
```

### Diagnose webhook gaps

```
Sent shows delivered messages, but our database lacks callbacks. Compare status, activity, webhook history, subscriptions, and endpoint responses before naming the cause.
```

### Build an escalation diagnosis

```
Analyze this multi-channel incident using explicit activity histories only. Reconcile all groups, quantify anomalies, rank hypotheses, and prepare one evidence-based escalation recommendation.
```

## Limitations

- Requires user-provided message, activity, or webhook evidence.
- Does not fetch live Sent account data or call external APIs.
- Cannot prove carrier filtering when the available cohort is small or incomplete.
- Provider code references can become outdated and should be checked against current documentation.

## Best Practices

- Define the cohort by time, profile, template, channel, country, and recipient segment before calculating rates.
- Use Sent message identifiers and explicit activity histories for transition analysis.
- Report counts with rates, retain unknown records, and end with one evidence-based verification step.

## Anti Patterns

- Do not infer missing lifecycle stages from a latest-status snapshot.
- Do not combine SMS, WhatsApp, and RCS before checking each channel separately.
- Do not label traffic as carrier filtered without a stable baseline and sufficient evidence.

## Security Audit

- Audited at: 2026-09-11T12:31:14.381\+00:00
- Summary: All 57 static findings are false positives caused by Markdown code formatting, documentation terminology, Python docstrings, and ordinary error handling. The script only reads a user-selected CSV or JSON file, computes local statistics, and prints results; no prompt injection, command execution, network access, or malicious intent was found.

## Stats

- Views: 0
- Downloads: 4
- Favorites: 0
- Popularity score: 0