ZensInk

SEO Toolkit for Indie Builders

CLI tools for keyword research, competitor analysis, and technical audits. Zero dependencies, pure Python.

English · 中文

---

Twenty CLI tools that cover the full SEO workflow — from discovering what people search, to clustering keywords by topic, to classifying search intent, to auditing your own site. No paid APIs. No Ahrefs. No subscriptions. Just free data sources wired together with Python.

Tool	What it does	API needed
keyword_research	Discover long-tail keywords via Google Autocomplete	None
keyword_cluster	Group keywords into semantic topic clusters	None
kd	Keyword difficulty score via SERP structure analysis	Free Serper key
kgr_auto	KGR opportunity scoring (competition vs volume)	Optional Bing key
keyword_volume	Real search volume via Bing Webmaster API	Free Bing key
brave_volume	Cross-check search demand via Brave SERP signals	Free Brave key
domain_rating	Real Ahrefs Domain Rating (0-100) via the free public API	Free Ahrefs key
search_intent	Classify keywords by search intent (info/commercial/transactional)	None
content_matrix	Prioritized content opportunity matrix	None
competitor_gap	Compare multiple competitor sitemaps, find content gaps	None
site_audit	Technical SEO + GEO audit: 30 checks covering links, meta, images, structured data, AI visibility	None
serp_intent	SERP-based intent analysis — reverse-engineer Google's actual ranking behavior	Free Serper key
onpage_audit	On-page quality scoring (7 dimensions, 0-100 per page)	None
geo_fanout	Query Fan-out content planning (GEO/AI search)	None
reddit_blueocean	Reddit blue-ocean keyword mining via Autocomplete	None
content_qc	Pre-publish gate for AI-citable content (14 weighted checks)	None
llms_gen	Generate llms.txt + llms-full.txt from a build dir or sitemap	None
ai_crawler_audit	robots.txt AI-crawler policy + llms.txt layer audit, AI-readiness score	None
rank_tracker	SQLite-backed keyword position history	Free Serper key
setup_gsc	One-time OAuth setup for Google Search Console	—
search_performance	Your site's real Google ranking data (GSC)	Free GSC OAuth

Why

Ahrefs costs $200/month. SEMrush costs $130/month. For indie builders who just need to find keywords worth writing about, that's overkill.

zens.ink wires together free public data sources — Google Autocomplete, Bing Webmaster Tools, Google Search Console, Brave Search, Ahrefs Domain Rating — with zero Python dependencies.

Install

Option A — pip (recommended)

pip install zens-ink

After install, the zens-ink command is available globally:

zens-ink --help

Option B — git clone

git clone https://github.com/ZensInk/zens-ink-seo-package.git
cd zens-ink-seo-package

Quick Start

# Discover keywords (with a-z long-tail expansion)
zens-ink keyword_research "tarot" --expand

# Check real search volume
zens-ink keyword_volume "tarot reading" --country us

# Keyword difficulty (SERP-based, with Chinese mode)
zens-ink kd "塔罗牌" --zh

# Compare 3 competitors at once
zens-ink competitor_gap \
  --url https://yoursite.com/sitemap.xml \
  --compare https://competitor-a.com/sitemap.xml https://competitor-b.com/sitemap.xml

# Audit your build for SEO issues
zens-ink site_audit --dist dist --sitemap dist/sitemap.xml

# Check your own Google search performance
zens-ink search_performance

Not installed? Use python3 -m instead

python3 -m zens_ink.keyword_research "tarot" --expand
python3 -m zens_ink.kd "tarot reading"
python3 -m zens_ink.site_audit --dist dist --sitemap dist/sitemap.xml

Use as Agent Skill

ZensInk works as an AI agent skill — let your AI assistant run SEO tools for you in plain language.

# Install for ClawHub / OpenClaw / Hermes compatible agents
npx skills add ZensInk/zens-ink-seo-package --skill zens-ink

Then just tell your AI: "find keywords for my tarot site" and it runs the tools for you. See SKILL.md for details.

Use as MCP Server (DSH / Claude / Codex)

ZensInk ships a built-in MCP stdio server — every tool becomes a native model tool (mcp__zensink__keyword_research, mcp__zensink__kd, ...). Pure stdlib, zero extra dependencies.

python3 -m zens_ink.mcp    # MCP stdio server on stdin/stdout

Add to DeepSeek Harness (DSH) — ~/.dsh/profiles/web/cordis.patch.yml:

- insert:
    - id: mcp-zensink
      name: '@deepseek-ai/dsh-mcp-client'
      config:
        serverName: zensink
        transport: stdio
        command: python3
        args: ['-m', 'zens_ink.mcp']
        toolCallTimeoutMs: 300000

For other MCP clients (Claude Desktop, Codex, ...), point the stdio command at python3 -m zens_ink.mcp. API keys are read from the package root .env as usual.

If the Pro package (zens_ink_pro) is installed alongside, its tools (winability, content_radar, competitor_radar, geo_score, geo_visibility, gap_deep, full_audit) are picked up automatically — the same server exposes everything available on that machine. OSS-only installs stay OSS-only.

Typical Workflow

keyword_research  →  find what people actually search
        ↓
keyword_volume    →  filter by real demand
        ↓
kd                →  pick keywords you can actually rank for
        ↓
competitor_gap    →  see what competitors already cover
        ↓
site_audit        →  make sure your pages are crawlable

Requirements

• Python 3.10+
• Zero dependencies — pure standard library
• Optional API keys (Bing, Serper, Brave, GSC) in .env — see .env.example

Documentation

• Changelog: CHANGELOG.md

• Full docs & case studies: zens.ink/docs

• Pro package (automated full-audit pipeline, HTML reports, GEO score): zens.ink

License

MIT — free for personal and commercial use.

---

Made by Jask