What does this skill do, and when should you use it?
Performance Monitor is an SEO/GEO evaluation skill with two modes. Report mode combines traffic, rankings, GEO/AI visibility, authority, backlinks, and content metrics into a period-over-period dashboard. Alert mode configures thresholds and notifications for ranking, traffic, technical, backlink, competitor, and AI-citation changes. It does not replace content-quality or domain-authority audits, and it does not calculate raw position-by-position ranking deltas.
It reads current and prior-period SEO/GEO metrics, baselines, normal volatility, critical keywords, and delivery preferences; with connectors, it can aggregate analytics, Search Console, SEO-tool, AI-monitoring, and crawler data. Its ledger workflow records and compares KPIs, while labeling figures as Measured, User-provided, or Estimated. Report mode produces a source-tagged multi-metric dashboard with changes, plausible explanations, recommendations, and a monitoring handoff summary. Alert mode produces named triggers, thresholds, priorities, response plans, and delivery routing.
- A marketing lead needs an executive SEO report comparing last month with the previous month, with source labels on traffic, ranking, and AI-citation metrics.
- An SEO team wants alerts for drops in organic traffic, keyword positions, indexed pages, crawl errors, or Core Web Vitals.
- A GEO owner needs to monitor AI citations for important queries and be notified when visibility declines.
- An agency needs a client-ready report covering authority, backlinks, content performance, and prioritized next actions.
How do you install this skill?
- Scripts or connectors were not executed, so availability of ledger, template references, and alert-configuration paths on the target host is unconfirmed.
- The two alert-threshold references conflict, including different traffic warning and critical values; one canonical rule should be selected and recorded before deployment.
- Alert routing may involve external effects such as email, SMS, or Slack; the supplied skill evidence does not show per-send confirmation, permission scoping, or rollback.
- The skill declares version 18.0.0 while SECURITY.md lists 17.x as the supported line; maintenance and security-support status should be reconciled.
- Shell / CLI
- Network access
- Local filesystem
Python 3 standard library
In Claude Code, run /plugin marketplace add aaron-he-zhu/aaron-marketing-skills, then /plugin install aaron-marketing@aaron. Other Agent Skills-compatible hosts can use npx skills add aaron-he-zhu/aaron-marketing-skills, or clone the repository with git clone https://github.com/aaron-he-zhu/aaron-marketing-skills. To retrieve only this skill, use npx skills add aaron-he-zhu/aaron-marketing-skills -s performance-monitor.
tmp="$(mktemp -d)"
git clone --depth 1 https://github.com/aaron-he-zhu/aaron-marketing-skills.git "$tmp"
mkdir -p ~/.claude/skills
cp -R "$tmp/seo-geo/evaluate/performance-monitor" ~/.claude/skills/
rm -rf "$tmp"Generated from the source repository and skill path; it copies only this skill's folder. If the author's install steps above differ, follow those first. To scope it to one project, replace ~/.claude/skills with that project's .claude/skills.
How do you use this skill?
Once installed, send your agent any of these to trigger it:
- Create an SEO performance report for [domain] for [time period].
- Set up SEO monitoring alerts for [domain].
The shortest invocation is performance-monitor <domain>. For reporting, prompt: “Create an SEO performance report for [domain] for [time period].” For monitoring, prompt: “Set up SEO monitoring alerts for [domain].” You can also pass --mode report or --mode alert; report mode needs a reporting and comparison period, while alert mode needs a baseline or normal-volatility reference.
What are this skill's strengths and limitations?
- Combines retrospective SEO/GEO reporting with forward-looking monitoring alerts.
- Covers traffic, rankings, GEO/AI visibility, authority, backlinks, and content performance.
- Requires source labels, period comparisons, and control-based readback decisions, helping separate evidence from speculation.
- Works with user-supplied exports when connectors are unavailable; the repository is licensed under Apache-2.0.
- Without connected data sources, users must provide analytics, Search Console, ranking, and KPI data.
- Reports require a defined reporting and comparison period; alert thresholds need a baseline or must be labeled Estimated when using defaults.
- It does not calculate CORE-EEAT or CITE scores and does not handle raw position-by-position ranking deltas.
- The supplied source does not document a standalone test suite or a platform-by-platform validation matrix for this individual skill.
How does this skill compare with similar options?
Side by side with related skills; every score comes from the same FSRS standard.
| Skill | FS score | Stars | Last updated | License |
|---|---|---|---|---|
| SEO Performance Monitor this page | 54 · Use with care | ★ 3k | 12d ago | Apache-2.0 |
| Rank Tracker | 52 · Use with care | ★ 3k | 12d ago | Apache-2.0 |
| GEO Client Report Generator (geo-report) | 48 · Use with care | ★ 11k | 21d ago | MIT |
| Blog Decay Analyzer | 44 · Not recommended | ★ 2.4k | 13d ago | MIT |
| SEO Drift Monitor | 49 · Use with care | ★ 19k | 22d ago | MIT |
Unlike the repository’s rank-tracker, this skill handles multi-metric SEO/GEO reports and threshold alerts; raw position-by-position ranking changes belong to rank-tracker. CORE-EEAT and CITE scoring are handled by content-quality-auditor and domain-authority-auditor respectively.
How did FollowSkills review this skill?
The skill separates report and alert modes, requires source labels, treats fetched content as untrusted data, and asks before writing results to memory. However, it does not fully specify least-privilege permissions, sensitive-data handling, confirmation for alert setup, or rollback; external connectors and possible email effects are mostly covered only by shared repository material, so 8 points are deducted.
Mode selection, missing-data handling, baseline gates, and completion criteria are fairly explicit, and the skill requires computed deltas rather than eyeballing. However, nothing was executed, script and reference availability were not reproduced from the supplied evidence, and the two threshold references contain material inconsistencies, so the static score is limited to 7.
Triggers, report/alert scenarios, non-fit scope, and Chinese-language use cases are clearly stated, with a pasted-export fallback when tools are unavailable. Input schemas, edge-case boundaries, host compatibility, and mainland-China reachability are not sufficiently evidenced, so 4 points are deducted.
The documentation includes Quick Start, a contract, data-source guidance, decision gates, templates, references, and next-skill routing; version and Apache-2.0 licensing are explicit. Maintenance ownership, update path, and changelog evidence are incomplete within the selected skill, and the repository security policy lists supported version 17.x while the skill declares 18.0.0, so 5 points are deducted.
The output structure, report sections, alert fields, owners, priorities, and response plans are well specified, so the core task appears achievable when data is supplied. There is no static evidence of a real directly usable output or execution, and threshold conflicts plus external-data dependence may require manual correction, so the score is capped conservatively at 6.
The skill requires source, date, measurement-status labels, control comparisons, and readback windows, providing some auditability. The supplied material contains no independently reproducible result, third-party corroboration, or committed key-path test evidence, so the score is 3.
Open a dimension to read why it scored that way
Evidence confidence：Low — Mostly static review, author material or a limited demo; useful for discovery, not high-risk decisions.
See the full review method →