An Ahrefs/Semrush-class SEO platform built for AI agents, not humans. Exposed as an MCP server.
36 tools with highly inconsistent quality. Most lack input schemas entirely (inferred from file structure, not visible in source). Descriptions range from adequate (seo_crawl_site: 'Crawl a website...') to vague (seo_audit_issues, seo_page_inspect: no descriptions shown). Only 3 tools have visible input schemas with type/description pairs. No output schemas documented. No error handling guidance. Tool names follow verb_noun pattern (good), but parameter documentation is sparse. Composition is reasonable (tools chain logically), but schema completeness is critical blocker.
Build a programmatic SEO publishing plan.
Check for thin content and duplication risks before publishing.
Discover programmatic SEO patterns in keywords.
Analyze visibility in AI search results.
Analyze crawl results for technical SEO issues.
Analyze backlink profile.
Clear local cache.
Show which data sources are configured and what unlocks the rest.
33 of 36 tools lack visible input schemas. Only seo_crawl_site and seo_keyword_ideas show parameter definitions with types and descriptions. Remaining tools have no documented input structure, violating pattern:tool and pattern:constrained-input. LLMs cannot infer valid parameters.
| Scored | Grade | Overall | Spec posture | Rubric |
|---|---|---|---|---|
| 2026-09-23 | F | 31 | 2026-07-28+ | v2 |
Cluster keywords by topic and intent.
Analyze competitor rankings and keywords.
Generate a content brief for a keyword.
Find content gaps vs competitors.
Score content against SEO best practices.
Compare two crawls to identify changes.
View crawl history for a site.
Crawl a website and return technical audit data.
Get domain authority metrics.
Find optimization opportunities from GSC data.
Get performance data from Google Search Console.
List Google Search Console properties.
Analyze internal linking structure.
Analyze keyword difficulty.
Generate keyword ideas from a seed term.
Get search volume and difficulty metrics for keywords.
Find backlink opportunities vs competitors.
Find internal linking opportunities.
Ranked, machine-applicable fixes with impact scores for SEO improvement.
Inspect a specific page from a crawl.
Get on-page optimization recommendations.
Analyze page speed and Core Web Vitals.
List configured projects.
Create or update a project configuration.
Track ranking changes over time.
Check rankings for keywords.
Get SERP results for a keyword.
Show API usage and quota information.
23 tools have no descriptions in the provided source. Tools like seo_audit_issues, seo_page_inspect, seo_keyword_metrics, seo_serp, seo_content_brief lack any explanation of purpose, when to call them, or what they return. Violates pattern:tool-description baseline (100% of A+ tools have docstrings).
No output schemas documented for any tool. LLMs cannot plan downstream calls or extract required fields. Violates pattern:tool baseline (100% of A+ tools have documented return types). No guidance on pagination, field structure, or error responses.
No error handling guidance. Tools that call external APIs (Ahrefs, Semrush, Google Search Console) can fail with provider-not-configured, rate limits, or network errors. No recovery hints, error classification, or actionable messages. Violates pattern:recovery-guide.
Tool annotations (readOnlyHint, destructiveHint, idempotentHint) not present. seo_project_set and seo_cache_clear are WRITE operations but lack destructiveHint. Violates current MCP spec (2026-07-28) tool annotation requirement for agent safety.