Install
$ agentstack add mcp-cociugv-yseo-pro-ai ✓ scanned · ✓ verified, works with Claude Code, Cursor, and more.
Security review
✓ PassedNo issues found. Passed automated security review. · v0.1.0 How review works →
- ✓ Prompt-injection patterns
- ✓ Secret / credential exfiltration
- ✓ Dangerous shell & filesystem operations
- ✓ Untrusted network calls
- ✓ Known-malicious package signatures
What it can access
- ✓ Network access No
- ✓ Filesystem access No
- ● Shell / process execution Used
- ● Environment & secrets Used
- ✓ Dynamic code execution No
From automated source analysis of v0.1.0. “Used” means the capability is present in the source — more access means more to trust, not that it’s unsafe.
Verified badge
Passed review? Show it. Paste this badge into your README, it links to the public security report.
Reliability & compatibility
Declared compatibility
Compatibility is declared by the source manifest. End-to-end runtime verification is coming, see below.
We're building live execution health for every listing: tool-call success rate, median latency, uptime, and last-checked timestamps, measured, not self-reported. It isn't live yet, so we don't show numbers we can't stand behind.
How agent discovery & health will work →About
ySEO-PRO-AI
The open-source SEO automation platform that doesn't just audit — it fixes.
11 real MCP tools for technical SEO, auto-fixing, multilingual intelligence, drift monitoring, AI search readiness, and competitive analysis. Works with Claude, Cursor, Codex, Gemini, Windsurf, and any MCP client.
[](LICENSE) [](https://modelcontextprotocol.io) [](#testing) [](https://python.org)
Why ySEO-PRO-AI?
| | Traditional SEO Tools | Audit-Only MCP Servers | ySEO-PRO-AI | |--|----------------------|------------------------|-----------------| | Audit | Reports in dashboard | Lists issues | Finds issues | | Fix | Manual work | Manual work | Auto-fixes | | AI Search | Not covered | Basic | AI Readiness Score | | Multilingual | Basic hreflang check | Not covered | 14-language intelligence | | Monitoring | Separate tool | Not covered | Built-in drift alerts | | Programmatic | Not covered | Not covered | City/vs/usecase generator | | Indexing | Manual submission | Not covered | IndexNow auto-ping | | Cost | $100-500/month | Free (limited) | Free + extensible |
Quick Start
Install as MCP Server (Claude / Cursor / Windsurf)
{
"mcpServers": {
"yseo-pro-ai": {
"command": "npx",
"args": ["-y", "yseo-pro-ai"]
}
}
}
Install from Source
git clone https://github.com/cociugv/ySEO-PRO-AI.git
cd ySEO-PRO-AI
npm install
Then ask your AI assistant: "Audit https://example.com for SEO issues and fix them"
MCP Tools (11 real implementations)
Audit
seo_audit_page— Full page audit (30+ checks, all modules)seo_check_indexability— Robots, meta robots, canonical, X-Robots-Tag
Auto-Fix
seo_fix_auto— Fix all fixable issues (with dry-run support)seo_fix_schema— Auto-detect page type + generate JSON-LDseo_fix_hreflang— Generate complete hreflang tag set
Analysis
seo_score_ai_readiness— AI Search Readiness Score (0-100)
Monitoring
seo_monitor_baseline— Capture SEO state snapshotseo_monitor_compare— Detect drift vs baseline
Competitor Intelligence
seo_competitor_compare— Side-by-side comparisonseo_backlink_opportunities— Find link building targets
Publishing
seo_indexnow_ping— Notify search engines instantly (batch support)
Architecture
┌─────────────────────────────────────────────────────────┐
│ MCP Clients │
│ Claude Desktop · Cursor · Codex · Gemini · Windsurf │
└────────────────────────┬────────────────────────────────┘
│ stdio (MCP Protocol)
┌────────────────────────▼────────────────────────────────┐
│ TypeScript MCP Server │
│ 11 tools · Zod validation · Deep adapter │
└────────────────────────┬────────────────────────────────┘
│ subprocess bridge
┌────────────────────────▼────────────────────────────────┐
│ Python Analysis Engine │
│ │
│ ┌──────────┐ ┌──────────┐ ┌──────────┐ ┌──────────┐ │
│ │Inspector │ │ Doctor │ │Polyglot │ │Sentinel │ │
│ │(audit) │ │(auto-fix)│ │(i18n) │ │(drift) │ │
│ └──────────┘ └──────────┘ └──────────┘ └──────────┘ │
│ ┌──────────┐ ┌──────────┐ ┌──────────┐ ┌──────────┐ │
│ │Architect │ │Publisher │ │ Radar │ │Citadel │ │
│ │(pages) │ │(publish) │ │(compete) │ │(schema) │ │
│ └──────────┘ └──────────┘ └──────────┘ └──────────┘ │
│ │
│ Pipeline: SCAN → DIAGNOSE → FIX → VERIFY → REPORT │
└─────────────────────────────────────────────────────────┘
Design Principles
- Fix, don't just audit — Every diagnosed issue has an auto-fix path
- MCP-native — First-class MCP server, works with any compatible client
- Zero-config start — Works out of the box, configure only what you need
- Multilingual by default — Built for international sites (14+ languages)
- AI-search aware — Optimizes for Google AI Overviews, ChatGPT, Perplexity
- Plugin-extensible — Add custom tools, integrations, and data sources
- Privacy-first — Runs locally, no data sent to third parties
Platform Compatibility
| Platform | Integration | Setup | |----------|-------------|-------| | Claude Desktop | MCP server (stdio) | Add to mcp.json | | Claude Code | MCP + Skills + Agents | Plugin auto-discovery | | Cursor | MCP server | Add to .cursor/mcp.json | | Kiro CLI | Steering rules + MCP | Open project in Kiro | | Windsurf | MCP server | Add to configuration | | Codex CLI | Python CLI | python -m src.ops.cli | | Gemini Code Assist | Python CLI | Direct execution | | n8n / Zapier | HTTP bridge | Via MCP HTTP transport | | Any MCP client | stdio transport | Standard MCP protocol |
CLI Usage
# Quick page scan
python -m src.ops.cli scan https://example.com
# Full audit with all modules
python -m src.ops.cli audit https://example.com
# Auto-fix (preview)
python -m src.ops.cli fix https://example.com --dry-run
# Auto-fix (apply)
python -m src.ops.cli fix https://example.com
# AI Search Readiness Score
python -m src.ops.cli ai-score https://example.com
Configuration
Edit config/default.yaml:
targets:
primary_domain: "example.com"
languages: [en, de, fr, es, ro]
modules:
doctor:
auto_fix: true
dry_run: false
sentinel:
snapshot_interval_hours: 24
publisher:
indexnow_key: "your-key"
Environment variables (.env):
INDEXNOW_KEY=your-key
PAGESPEED_API_KEY=optional
GSC_CREDENTIALS_JSON=optional
Modules
| Module | Purpose | Key Feature | |--------|---------|-------------| | Inspector | Technical SEO scanner | 30+ diagnostic checks | | Doctor | Auto-fix engine | Generates fixes, not just reports | | Polyglot | Multilingual SEO | Hreflang audit + generation | | Sentinel | Drift monitoring | Snapshot comparison + alerts | | Architect | Programmatic SEO | City/comparison/usecase pages | | Publisher | Content publishing | Blog API + IndexNow ping | | Radar | Competitor intel | Compare + backlink opportunities | | Citadel | Schema + AI readiness | Auto-inject + scoring |
Testing
# Run full test suite (113 tests)
python -m pytest tests/ -v
# Run only benchmark fixtures
python -m pytest tests/test_benchmark.py -v
# Run specific module tests
python -m pytest tests/test_remediation.py -v
Test coverage:
- Parser: 21 tests (extraction, broken HTML, serialization)
- Pipeline: 13 tests (stages, execution, scoring, hooks)
- Doctor: 9 tests (fix status honesty, generation, artifacts)
- Modules: 22 tests (hreflang, drift, architect, radar, citadel)
- Operations: 8 tests (deep module, typed records)
- Remediation: 18 tests (state machine, file adapter, integration)
- Benchmark: 22 tests (precision/recall against HTML fixtures)
Contributing
Contributions welcome! See [CONTRIBUTING.md](CONTRIBUTING.md) for guidelines.
Key areas:
- New MCP tools
- Additional fix prescriptions
- Integration plugins (Ahrefs, Semrush, etc.)
- Translations for multilingual content templates
- Documentation improvements
License
MIT License — Vadim Cociug
Links
- [Documentation](docs/)
- [Installation Guide](docs/INSTALLATION.md)
- [Tools Reference](docs/TOOLS.md)
- [Architecture](docs/ARCHITECTURE.md)
- [Plugin Development](docs/PLUGINS.md)
Source & license
This open-source MCP server is cataloged on AgentStack and links to its original source — we do not rehost the code.
- Author: cociugv
- Source: cociugv/ySEO-PRO-AI
- License: MIT
Install and usage instructions live in the source repository linked above.
Reviews
No reviews yet, be the first.
Write a review
Versions
- v0.1.0 Imported from the upstream source.