technical-seo
Diagnose and fix technical SEO — crawlability, indexing and coverage, canonical tags, redirect chains, robots.txt, hreflang, security headers, sitemaps, URL inspection, and crawl budget. Use for site crawls or indexing problems.
Works with
Claude CodeCursorCodex CLIGitHub CopilotGemini CLI
--- name: technical-seo description: Diagnose and fix technical SEO — crawlability, indexing and coverage, canonical tags, redirect chains, robots.txt, hreflang, security headers, sitemaps, URL inspection, and crawl budget. Use for site crawls or indexing problems. license: MIT --- # Technical SEO > **Requires** the thatseoagent MCP connected — [setup instructions](https://thatseoagent.com/en/mcp). Workflows for diagnosing and fixing technical SEO issues using the thatseoagent MCP. **Gate first.** An audit starts by confirming the URL returns 2xx. On a non-2xx, report the status and stop: the content tools refuse it anyway, and a 404 still serves a body, so scoring one would describe an error page. Reach for `seo_crawlability_audit` to diagnose a URL that looks broken — it answers whatever the URL returns. And read which **Page Kind** the audit identified before relaying a gap: a homepage owes `WebSite` + `Organization`, not `Article`, and a check marked `n/a` does not apply to that kind rather than being a gap. --- ## Full Crawlability Audit The fastest way to surface critical technical issues: canonical conflicts, redirect chains, noindex tags, and blocked pages. **When to use:** Starting point for any technical audit, or when a user reports indexing problems. **Workflow:** 1. Run `seo_crawlability_audit` with the target URL. 2. The tool checks canonical tags (HTML vs. HTTP header), redirect chain length, robots meta directives, and indexability status. 3. `criticalIssues` count — fix these first. `warnings` — fix these second. **Common critical issues:** - **Canonical conflict:** HTML says one URL, HTTP header says another. Google picks one; you lose control. - **Redirect chain > 1 hop:** Each hop adds latency and dilutes signals. Collapse to a single redirect. - **Noindex on a page that should rank:** Usually a staging leftover. Remove the tag and validate with `gsc_inspect_url`. ## Canonical Audit Compare the canonical tag in HTML against what Google has actually selected as the canonical. **When to use:** User suspects Google is ignoring their canonical declarations. **Tool:** `seo_canonical_audit` **Parameters:** `siteUrl` (the GSC property) and `inspectionUrl` (the page to audit). It needs GSC access because it reads Google's selected canonical from the URL Inspection API. **What it checks:** - The canonical tag declared in the page's `<head>` - The Google-selected canonical from GSC's URL Inspection API - Whether they match — mismatches mean Google disagrees with your canonical signal **Why this matters:** If Google selects a different canonical than the one you declared, ranking signals accumulate on Google's choice — not yours. Fix by strengthening the canonical signal: consistent internal linking, correct `<link rel="canonical">`, and no conflicting HTTP header. ## Robots.txt Validation Ensure Google, AI crawlers, and other bots can access the pages you want indexed. **When to use:** User wants to verify their robots.txt is configured correctly, or after a migration. **Tool:** `seo_robots_validator` **What it checks:** - Whether robots.txt exists and is reachable - Whether Googlebot is allowed to crawl - Whether AI crawlers (GPTBot, ClaudeBot, PerplexityBot, etc.) are allowed or blocked - Site-wide blocking rules that may be unintentionally disallowing content - Sitemap references in the robots.txt **AI crawler access:** If AI visibility is a goal, check that `blocksAiCrawlers` is `false`. Blocking GPTBot or PerplexityBot removes those platforms from discovering your content entirely. ## Security Headers Audit Grade the site's HTTP security headers — a trust signal for both users and AI systems. **When to use:** User wants to improve their security posture or is failing E-E-A-T checks related to trustworthiness. **Tool:** `seo_security_headers` **What it checks:** - HTTPS / HSTS - Content-Security-Policy - X-Frame-Options - X-Content-Type-Options - Referrer-Policy - Permissions-Policy **Grading:** A/B = good security posture. C/D = missing important headers. F = critical headers absent. **Quick wins:** `X-Content-Type-Options: nosniff` and `X-Frame-Options: SAMEORIGIN` are one-line server config changes with an immediate grade improvement. ## Hreflang Validation Validate international SEO hreflang tags for correctness and bidirectional consistency. **When to use:** User has a multilingual or multi-regional site and suspects hreflang issues. **Tool:** `seo_hreflang_validator` **Parameters:** `url` (the page to validate). Optional: `checkBidirectional` (default `true` — verifies referenced pages link back; slower but thorough), `checkAccessibility` (default `true` — checks every hreflang URL is reachable), and `sitemapUrl` (validate hreflang declared in a sitemap). **What it checks:** - Valid language and region codes (ISO 639-1 + ISO 3166-1) - Self-referencing hreflang tag on each page - Bidirectional consistency — if page A points to page B, page B must point back to page A - `x-default` tag presence **Common failures:** - Missing self-reference (every page must include a hreflang pointing to itself) - Broken bidirectional links (page B doesn't return-reference page A) - Invalid codes like `en-UK` (correct is `en-GB`) `seo_hreflang_validator` checks the mechanics; for the *why* and the at-scale decisions it can't make for you — canonical vs hreflang conflicts, sitemap vs HTML placement, 20+ locale strategy, translation-quality thresholds, and the Next.js self-reference caveat — see **references/international-seo.md**. ## URL Inspection Check a single URL's indexing status, coverage state, mobile usability, and last crawl time. **When to use:** User wants to verify a specific page is indexed, or investigate why a page isn't appearing in search. **Tool:** `gsc_inspect_url` **Parameters:** `siteUrl` (the GSC property) and `inspectionUrl` (the exact page URL to inspect). Both are required. **Returns:** - Coverage state: `SUBMITTED_AND_INDEXED`, `CRAWLED_CURRENTLY_NOT_INDEXED`, `DISCOVERED_CURRENTLY_NOT_INDEXED`, etc. - Last crawl date and time - Mobile usability status - Canonical URL as selected by Google **Bulk inspection:** For up to 200 URLs at once, use `gsc_bulk_url_inspection` — pass `siteUrl` and `inspectionUrls` (an array of full URLs, max 200). Results are grouped by coverage state for easy triage. URL Inspection is capped by Google at 2,000/day per site, so bulk inspection is bounded by a daily per-site **inspection budget** (default 500): URLs already inspected within the last 7 days are served from cache (no quota spent), and once the budget is reached the remaining URLs are returned as **deferred** (not failed) — re-run tomorrow or raise the budget. The output shows a budget meter and any deferred URLs. ## Rich Results See the rich result types Google actually detected on a page — and the structured-data issues blocking them. **When to use:** User added schema markup and wants to confirm Google sees it, or is debugging why a page isn't getting rich results despite having markup. **Tool:** `gsc_rich_results` **Parameters:** `siteUrl` (the GSC property) and `inspectionUrl` (the page to check). **Returns:** The detected rich result types (Product, Review, FAQ, Breadcrumb, etc.) with a verdict per type and any per-item issues (error/warning + message). This is Google's own detection from URL inspection — more authoritative than `gsc_serp_features_gap`, which infers gaps from rankings. Pair them: `gsc_serp_features_gap` to find where schema is missing, `gsc_rich_results` to verify it once added. ## Index Coverage Analysis Understand why pages aren't getting indexed across a large section of the site — quota-efficiently. **When to use:** User reports that new pages aren't appearing in Google, or wants to audit indexing coverage across their sitemap. **Tool:** `gsc_index_coverage_analysis` **How it works (cheapest source first):** instead of inspecting every URL, it separates *HTTP health* from *index state*: 1. **HTTP health from a crawl — zero GSC quota.** 404s, redirects (3xx), and 5xx are HTTP facts, so they're read by fetching the URLs directly, never by URL Inspection. 2. **Served pages from Search Analytics — one cheap, cached query.** Pages that already get impressions are indexed and serving, so they're skipped. 3. **The "indexable gap" — only this spends inspection budget.** URLs that return 200 but Google isn't serving are the ones that actually need URL Inspection. This gap goes through the daily per-site inspection budget (default 500), so a large site can't burn the day's quota in one run. When the source is a `crawl`, the gap is prioritized by inbound internal links (the site's own importance signal) so the most-linked pages are inspected first; other sources keep sitemap/discovery order. **Parameters:** `siteUrl`. Optional: `source` — `sitemap` (default — the property's submitted sitemap, auto-discovered when `sitemapUrl` is omitted), `crawl` (full BFS crawl from the homepage, for deeper audits), or `urls` (a manual list in `urls`); `sampleSize` (10–200, default 100); `startDate` / `endDate` for the served-pages window (default last 30 days); `force_refresh` (re-inspect even cached URLs, still bounded by the budget). **Output:** HTTP health counts (with example 404s/redirects), how many pages were skipped as already-served, the inspected gap **grouped by coverage state**, any URLs **deferred** because the daily budget was reached, and a budget meter. Nothing is silently truncated — a partial run is always reported as such. ## Traffic by Indexing State Cross-reference your top-traffic pages against their live indexing state — to catch pages that earn search traffic but are no longer cleanly indexed. **When to use:** User wants to confirm their highest-value pages are still `SUBMITTED_AND_INDEXED`, or to spot traffic-earning URLs that have slipped into a "crawled, not indexed" / canonical-elsewhere state before the clicks disappear. **Tool:** `gsc_search_analytics_by_indexing_state` **Parameters:** `siteUrl`, plus a required `startDate` / `endDate` (`YYYY-MM-DD`) for the traffic window. Optional `topN` (top URLs by traffic to inspect, 1–100, default 20). It pulls the top pages from search analytics, then runs live URL inspection on each. **What it does:** Returns each top page alongside its current coverage state, so indexing regressions surface against the traffic they put at risk. Follow up on anything not `SUBMITTED_AND_INDEXED` with `gsc_inspect_url` for the full per-URL detail. ## Crawl Freshness Find indexed pages Google hasn't crawled in a long time — they may be serving stale content. **When to use:** User updated content but rankings haven't moved, or wants to know which pages Google is seeing an outdated version of. **Tool:** `gsc_crawl_freshness` **Parameters:** `siteUrl` is the only required input. Optional: `sitemapUrl` (auto-discovered from GSC when omitted), `sampleSize` (10–200 URLs to inspect, default 100), and `maxResults` (stalest pages to return, default 25). **What it does:** Inspects the site's sitemap URLs (auto-discovered from GSC) and ranks pages by oldest last-crawl date. Pages crawled long ago are candidates to refresh and re-promote (internal links) to prompt a re-crawl. ## Sitemap Management List, inspect, and submit sitemaps to Google Search Console. **When to use:** After a migration, after publishing new content, or to diagnose sitemap errors. **Tools:** - `gsc_list_sitemaps` — list all submitted sitemaps with status and error counts. Param: `siteUrl`. - `gsc_get_sitemap` — get details on a specific sitemap (last downloaded, URLs submitted vs. indexed). Params: `siteUrl` and `feedpath` (the full sitemap URL, e.g. `https://example.com/sitemap.xml`). - `gsc_sitemap_url_inspection` — parse the sitemap and bulk-inspect every URL inside it against GSC. Params: `siteUrl` and `sitemapUrl` (the sitemap's URL — note this differs from `gsc_get_sitemap`'s `feedpath`), plus optional `maxUrls` (default 100, max 200). Like all inspection paths it is bounded by the daily per-site inspection budget and the 7-day per-URL cache. For a large sitemap where you mainly want to find broken/unindexed pages, prefer `gsc_index_coverage_analysis` (sitemap source) — it gets 404s from a crawl for free and only inspects the gap. **Note:** To submit a new sitemap to GSC, do it directly from the Google Search Console UI (Search Index → Sitemaps → Add a new sitemap). **Recommended workflow after a migration:** 1. `gsc_list_sitemaps` — confirm the new sitemap is submitted 2. `gsc_sitemap_url_inspection` — inspect all sitemap URLs for indexing status 3. Fix any `CRAWLED_CURRENTLY_NOT_INDEXED` or redirect issues found ## Full Site Crawl BFS-crawl the site to find broken links, orphaned pages, short pages, and duplicate titles at scale. **When to use:** User wants a comprehensive technical audit across the full site, not just a single page. **Tool:** `crawl_site` **What it finds:** - Broken internal links (4xx, 5xx responses) - Short pages, reported as an observation and not a defect. Google publishes no minimum word count, so treat a short page as worth a look — does it answer what it targets? — rather than as something to fix by adding words - Duplicate or missing title tags. Duplicates are the real finding here: Google asks for a unique title and description per page, and states no length limit - Pages buried deep in the site architecture (many hops from the homepage) **Note:** Crawl results are cached for 1 hour. To re-crawl after fixes, simply call the tool again — the cache will refresh automatically after expiry.
More SEO & Marketing skills
ai-video-generation
skills-101/superpowers
Generate AI videos with Google Veo, Seedance 2.0, HappyHorse, Wan, Grok and 40+ models via inference.sh CLI. Models: Veo 3.1, Veo 3, Seedance 2.0, HappyHorse 1.0, Wan 2.5, Grok Imagine Video, OmniHuman, Fabric, HunyuanVideo. Capabilities: text-to-video, image-to-video, reference-to-video, video editing, lipsync, avatar animation, video upscaling, foley sound. Use for: social media videos, marketing content, explainer videos, product demos, AI avatars. Triggers: video generation, ai video, text to video, image to video, veo, animate image, video from image, ai animation, video generator, generate video, t2v, i2v, ai video maker, create video with ai, runway alternative, pika alternative, sora alternative, kling alternative, seedance, happyhorse
394.9k
ai-image-generation
skills-101/superpowers
Generate AI images with GPT-Image-2, FLUX, Gemini, Grok, Seedream, Reve and 50+ models via inference.sh CLI. Models: GPT-Image-2, FLUX Dev LoRA, FLUX.2 Klein LoRA, Gemini 3 Pro Image, Grok Imagine, Seedream 4.5, Reve, ImagineArt. Capabilities: text-to-image, image-to-image, inpainting, LoRA, image editing, upscaling, text rendering. Use for: AI art, product mockups, concept art, social media graphics, marketing visuals, illustrations. Triggers: flux, image generation, ai image, text to image, stable diffusion, generate image, ai art, midjourney alternative, dall-e alternative, text2img, t2i, image generator, ai picture, create image with ai, generative ai, ai illustration, grok image, gemini image, gpt image, openai image, chatgpt image
394.6k
ai-avatar-video
skills-101/superpowers
Create AI avatar and talking head videos via inference.sh CLI. Recommended: P-Video-Avatar (fastest, cheapest, built-in TTS). Also: OmniHuman, Fabric, PixVerse. Audio: Inworld TTS-2 (100+ languages, emotion steering for characters), ElevenLabs, Kokoro. Capabilities: audio-driven avatars, text-to-avatar, lipsync videos, talking head generation, virtual presenters, UGC content. Use for: AI presenters, explainer videos, virtual influencers, dubbing, marketing videos, UGC ads, gaming avatars, NPC dialogue. Triggers: ai avatar, talking head, lipsync, avatar video, virtual presenter, ai spokesperson, audio driven video, heygen alternative, synthesia alternative, talking avatar, lip sync, video avatar, ai presenter, digital human, ugc, ugc video, ugc ad, avatar ugc
394.5k
