Skip to content

Releases

seo-audit v1.0.0

seo-audit v1.0.0 Latest
Latest

Choose a tag to compare

View all tags
@Haniubub Haniubub released this 30 Aug 01:17
v1.0.0
f0f87e7

What's this?

A native, self-contained local & technical SEO audit toolkit for DeepSeek Harness — deterministic Python measurement + LLM judgment, in one weighted, prioritised report. No SaaS, no plugin marketplace, no per-site pricing.

Highlights

  • Two-layer engine — deterministic measurement (specialists + 53 curated scripts, $0 in LLM tokens) plus LLM judgment (24 sub-skills + 18 specialist agents).
  • Weighted SEO Health Score — Technical 22% · Content/E-E-A-T 23% · On-Page 20% · Schema 10% · Performance 10% · AI-Readiness 10% · Images 5%.
  • Gated multi-agent fan-out — detects the business type, then spawns only the relevant sub-agents (never all 18), with credential gating.
  • Primary-source grounded — Google Search Essentials, helpful-content/E-E-A-T, Core Web Vitals, AI Optimization guidance.
  • Cost transparency — ~$0.04 per audit on DeepSeek vs ~$0.45–$1.13 on Claude (12–30× cheaper). The measurement layer is pure local Python ($0 in LLM tokens).

Setup

cd seo-toolkit
./setup.sh      # installs workspace-local pylibs/ + Playwright Chromium, then a smoke test
./seo audit https://example.com

./setup.sh now bootstraps its own workspace-local pylibs/ (pinned deps), so a fresh clone is fully self-contained — no dependence on globally installed packages.

Docs

  • README (English + 简体中文), architecture, ./seo doctor
  • docs/INSTALL.md — 3-step install
  • docs/TUTORIAL.md — end-to-end audit in 5 minutes
  • docs/SHARE.md + docs/POSTS.md — share copy
  • pdf/google-seo-reference.md — curated primary-source reference

License

MIT. A native re-implementation/port of AgriciDaniel/claude-seo v2.2.5 — upstream MIT notice preserved; see Attribution & License. Individual extension integrations (Google APIs, DataForSEO, Firecrawl, Ahrefs, Bing, Banana) require their own credentials.

Assets 2
Loading