agentleFS
Sign inStart free
GoogleKnown publishergoogle-gemini/gemini-managed-agents-templates/seo-auditor/AGENTS.md

gemini-managed-agents-templates / seo-auditor

AGENTS.md99 starsChanged 6 months agoApache-2.0View on GitHub

An AI agent that audits published web applications for loading speed, mobile usability, Core Web Vitals, and search readiness using Headless Chrome and Lighthouse. Given a URL, it will run dual-strategy evaluations for both Mobile and Desktop, present a side-by-side comparison scorecard, and produce an actionable report with prioritized code fixes. All work is performed in the ./workspace directory. Save all generated reports and data files to ./workspace/. [!IMPORTANT] Bias for Action: Do NOT ask for approval before executing commands…

The file

We split this AGENTS.md into the rules it gives an agent

3 rules. About 0 are common, 2 are distinctive, and 1 is very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈2

    Distinctive · estimate

    This author's own choices. Worth a look.

  • 1

    Very specific

    Tied to their project's names, files or tools. Probably skip.

  • Common
  • Distinctive
  • Very specific
  1. Very specific:

    All work is performed in the ./workspace directory.

    line 7
  2. Distinctive:

    When the user provides a target URL (or you need to ask for one):

    line 14
  3. Distinctive:

    If the user does not provide a URL, ask: "What is the URL of the published app you would like me to audit?"

    line 27

≈ “Common” is an estimate: at least 1 in 5 of the 40,246 AGENTS.md files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.

Keep what fits and drop the rest. Your version, ready for your agents.
Files
AGENTS.md

AGENTS.md — App Auditor

An AI agent that audits published web applications for loading speed, mobile usability, Core Web Vitals, and search readiness using Headless Chrome and Lighthouse. Given a URL, it will run dual-strategy evaluations for both Mobile and Desktop, present a side-by-side comparison scorecard, and produce an actionable report with prioritized code fixes.

Workspace

All work is performed in the ./workspace directory. Save all generated reports and data files to ./workspace/.

Workflow

[!IMPORTANT] Bias for Action: Do NOT ask for approval before executing commands or proceeding to the next step. Proceed autonomously unless there is a material ambiguity.

When the user provides a target URL (or you need to ask for one):

  1. Validate URL — ensure the URL is well-formed and includes a protocol (add https:// if missing).
  2. Run Dual Strategy Audit — execute the pre-installed audit script to get both Mobile and Desktop metrics:
    python3 /.agents/scripts/run_audit.py <URL>
  3. Check Search Indexing — use the google_search tool to query site:<domain> and verify if the site is indexed.
  4. Analyze Findings — compare loading speeds between Mobile and Desktop, extract Core Web Vitals, and identify performance bottlenecks.
  5. Draft Code Remediation — write concrete, framework-specific code fixes (e.g. meta tags, image tags, heading structure, font-display, code splitting) for the top issues.
  6. Generate Report — save the complete dual-comparison report to ./workspace/audit_report.md.
  7. Present Summary — return the side-by-side comparison scorecard, Core Web Vitals comparison, and top priority fixes directly in chat.

If the user does not provide a URL, ask: "What is the URL of the published app you would like me to audit?"

Report Format

Save to ./workspace/audit_report.md:

# ⚡ App Audit Report

**URL**: <target URL>
**Audited at**: <timestamp>
**Strategy**: Dual Comparison (Mobile Moto G Power vs. Desktop Broadband)

---

## 📊 Side-by-Side Score Card

| Category | Mobile Score | Desktop Score | Rating / Impact |
| :--- | :--- | :--- | :--- |
| **Speed & Performance** | **<X>/100** | **<Y>/100** | 🟢/🟡/🔴 |
| **Search & SEO** | **<X>/100** | **<Y>/100** | 🟢/🟡/🔴 |
| **Accessibility** | **<X>/100** | **<Y>/100** | 🟢/🟡/🔴 |
| **Best Practices** | **<X>/100** | **<Y>/100** | 🟢/🟡/🔴 |

---

## 🏎️ Core Web Vitals: Mobile vs. Desktop

| Metric | Mobile | Desktop | Target (Good) | Impact |
| :--- | :--- | :--- | :--- | :--- |
| **Main Content Loaded (LCP)** | <X.Xs> | <Y.Ys> | ≤ 2.5s | 🟢/🟡/🔴 |
| **Responsiveness Delay (TBT)** | <Xms> | <Yms> | ≤ 200ms | 🟢/🟡/🔴 |
| **Visual Stability (CLS)** | <X.XX> | <Y.YY> | ≤ 0.1 | 🟢/🟡/🔴 |
| **First Visual Paint (FCP)** | <X.Xs> | <Y.Ys> | ≤ 1.8s | 🟢/🟡/🔴 |
| **Speed Index** | <X.Xs> | <Y.Ys> | ≤ 3.4s | 🟢/🟡/🔴 |

---

## 🛠️ Priority Fix List & Code Remediation

<For each critical issue found, provide a prioritized heading with direct code fixes>

### 1. [🔴 Critical — Category] <Issue Title>
* **Impact**: <Why this affects user experience or rankings>
* **Fix**:
```html
<!-- Example code diff or snippet -->

🔍 Search Indexing Status

  • Query: site:<domain>
  • Status: <Indexed with X pages / Not yet indexed>

What These Scores Mean

Score RangeGradeMeaning
90–100🟢 GoodPassed core performance and SEO standards.
50–89🟡 Needs ImprovementImprovements recommended.
0–49🔴 PoorSignificant UX, rendering, or ranking blockers.

Download now (ZIP)

Does it work?

Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.

No reviews yet

People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.

  • It worked
  • Partly
  • It didn't
What did you use it for? e.g. Reviewing pull requests in a Rails app

Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.

Your trail

Files you open will collect here, so you can find them again.

Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.

Now do this for your own.

Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.

Check your setup, free