What it does
Pulls the month's key metrics, sessions, top posts, top referral sources, and conversions if tracked, and writes a short narrative report instead of a raw metrics dump, calling out what changed versus the prior month and why it likely matters. Ends with 2-3 concrete suggestions for the next month based on what performed well or poorly.
Install
This skill isn't packaged in an installable registry. It's a plain SKILL.md file you can copy directly. Copy it from the SKILL.md tab and save it at the path for your assistant:
Use it in your AI assistant
Claude Code
.claude/skills/monthly-report-writer/ (or ~/.claude/skills/monthly-report-writer/ for all projects)
GitHub Copilot (VS Code)
.github/skills/monthly-report-writer/
OpenAI Codex
.codex/skills/monthly-report-writer/
Cursor
.cursor/skills/monthly-report-writer/ (or ~/.cursor/skills/monthly-report-writer/ globally)
Compatibility
Requires a connected analytics source, such as Google Analytics or Jetpack Stats; works on any WordPress site with that integration.
The complete SKILL.md for this skill:
---
name: monthly-report-writer
description: Use when a user wants a monthly traffic and content performance report compiled, or asks for a "monthly recap" or "performance summary".
---
# Monthly Performance Report Writer
## When to use
It's the end of the month and the user wants a readable performance report built from the actual metrics, not just a table of numbers.
## Procedure
1. Pull the month's core metrics: total sessions, top 5 posts by traffic, top referral sources, and any tracked conversions.
2. Compare each metric against the prior month and note the direction and rough size of the change.
3. Write a short narrative summary, not a table dump, explaining the most notable change and a plausible reason for it.
4. Highlight the top-performing post or channel and note what made it work, if that's inferable from topic, timing, or promotion.
5. Close with 2-3 specific, actionable suggestions for next month tied directly to what the data showed.
## Guardrails
- Don't present monthly noise as a trend, a single month's dip or spike needs at least a second data point before calling it a pattern.
- Keep suggestions grounded in the actual data pulled, not generic best-practice advice unconnected to this site's numbers.