Skip to main content
Glama

⭐ If this agent-first tool helps your workflow, please star the repo. Stars make this tooling easier for other builders to discover and help Delx keep shipping open infrastructure. 🧱 Part of the Delx agent stack — 15 open-source MCP servers across body, reach and coordination.


Agent-first SEO scoring, search-intent detection and opportunity prioritization. It packages the useful parts of a production content pipeline into a clean local CLI plus an optional MCP server for Codex, Claude, Cursor, Hermes, OpenClaw and other agent runtimes.

Use it when an agent needs deterministic SEO checks before rewriting, refreshing or publishing content.

What It Does

  • Classifies search intent: informational, navigational, transactional and commercial investigation

  • Scores markdown articles for agent-readable SEO gaps

  • Prioritizes GSC-style opportunities by impressions, position, CTR gap, conversions and commercial value

  • Exposes manifest, connection_status and privacy_audit surfaces before content tools

  • Runs locally by default with no required API keys

Related MCP server: GeoRanker MCP Server

Install

pipx install agent-seo-engine

With MCP support:

pipx install "agent-seo-engine[mcp]"

Published on PyPI: agent-seo-engine. Release automation uses PyPI Trusted Publishing, so GitHub Actions can publish future versions without long-lived PyPI tokens. See docs/pypi-publishing.md.

CLI

agent-seo-engine manifest --client codex
agent-seo-engine doctor
agent-seo-engine privacy-audit
agent-seo-engine intent "best ai agent framework"
agent-seo-engine score --file examples/article.md --primary-keyword "ai agent testing"
agent-seo-engine opportunity --impressions 4200 --clicks 80 --position 12.4 --commercial-intent 0.8
agent-seo-engine image-alt --file page.html

All commands return structured JSON by default. Use --format markdown for human review.

MCP

agent-seo-mcp

Hermes-style config:

mcp_servers:
  agent_seo:
    command: agent-seo-mcp
    args: []
    sampling:
      enabled: false

Recommended first calls:

  1. agent_seo_connection_status

  2. agent_seo_privacy_audit

  3. agent_seo_score_content

Agent Surfaces

Tool

Purpose

agent_seo_manifest

Install/runtime guidance for agent clients

agent_seo_connection_status

Local/offline readiness and optional integration status

agent_seo_privacy_audit

Draft, analytics and credential boundaries

agent_seo_detect_intent

Search intent classification

agent_seo_score_content

Markdown quality checks with exact recommendations

agent_seo_prioritize_opportunity

GSC-style opportunity scoring

agent_seo_check_image_alt

Image alt-attribute coverage audit for HTML

Copy-Paste Agent Prompt

Use agent-seo-engine. First call agent_seo_connection_status and agent_seo_privacy_audit.
Score the draft, then propose only edits tied to failed checks or high-impact opportunities.

Agent Contract

Agents should not guess whether a draft is ready. They should call the scoring tool, read exact failed checks, then propose focused edits. The engine is intentionally deterministic and local so repeated agent runs can compare output over time.

Development

python3 -m venv .venv
. .venv/bin/activate
pip install -e ".[dev]"
pytest
python -m compileall -q src

šŸ“§ Contact & Support

Available Tools

7 tools
agent_seo_check_image_altD

Inputs: html, file_path.

agent_seo_connection_statusD

Inputs: no named parameters.

agent_seo_detect_intentD

Inputs: keyword (required).

agent_seo_manifestD

Inputs: client.

agent_seo_prioritize_opportunityD

Inputs: clicks, position, conversions, impressions, +1 more.

agent_seo_privacy_auditD

Inputs: no named parameters.

agent_seo_score_contentD

Inputs: markdown, file_path, primary_keyword.

Tool Schema Changelog

Recent tool additions, removals, and schema changes observed during successful MCP inspections.

Loading tool schema changelog…

    TDQS

    D1.9/5.0

    Scored across 7 tools

    Disambiguation5/5

    Each tool targets a distinct SEO function (e.g., image alt, intent detection, privacy audit), with no apparent overlap in purpose.

    Naming Consistency3/5

    While all tools share the 'agent_seo_' prefix, some follow a verb_noun pattern (check_image_alt, detect_intent, prioritize_opportunity, score_content) while others are plain nouns (connection_status, manifest, privacy_audit), creating inconsistency.

    Tool Count5/5

    Seven tools is a reasonable number for a specialized SEO assistant, providing focused functionality without being overly numerous or sparse.

    Completeness3/5

    The set covers important SEO checks (alt text, content score, privacy) but lacks common tasks like keyword research, meta tag optimization, or sitemap management, indicating notable gaps.

    Maintenance

    ActivityMaintained
    ResponsivenessOutdated data

    Related MCP Connectors

    Related MCP Servers

    • A
      license
      Not graded
      quality
      D
      maintenance
      Agent-first SEO toolkit with 24 MCP tools for keyword research, rank tracking, site audits up to 50k pages, competitor analysis, content gap detection, domain reputation, backlink intelligence, Google Search Console integration, and AI-powered strategy generation with Claude, GPT, and Ollama. SQLite-backed and bring-your-own-key.
      MIT
    • A
      license
      Not graded
      quality
      D
      maintenance
      An agent-friendly MCP server for the GeoRanker High-Volume API, enabling SEO rank tracking and keyword management through natural language.
      29 npm
      1
      MIT
    • A
      license
      A
      quality
      B
      maintenance
      AutomateLab AI SEO Magic Button runs a whole-site AEO/GEO audit and generates a prioritized plan an agent can execute. It combines AI-SEO and Citation Intelligence and ships as a skill, plugin, and MCP server. Product and documentation: https://automatelab.tech/products/skills-and-plugins/ai-seo-magic-button/
      1
      2
      MIT
    • A
      license
      A
      quality
      D
      maintenance
      Enables AI agents to perform comprehensive SEO audits on web pages, including meta tags, headings, links, images, performance, and more, via a CLI or MCP server.
      18
      2
      MIT