anymorph 0.4.0 → 0.5.0

package/dist/skillpacks/geo/skills/seo-opportunity-audit/references/action-contract.md

@@ -1,62 +0,0 @@
-# SEO Action Contract
-
-Use this contract for all SEO diagnosis and opportunity outputs.
-
-## Finding
-
-```json
-{
-  "id": "stable-local-id",
-  "sourceSkill": "seo-technical-diagnosis",
-  "findingType": "crawlability",
-  "target": {
-    "url": "https://example.com/page",
-    "pageId": "optional",
-    "intentId": "optional",
-    "queryCluster": "optional"
-  },
-  "summary": "Concise finding",
-  "evidence": [
-    {
-      "type": "crawl|html|rendered_html|dfs|cms|serp|firecrawl|product",
-      "source": "tool, file, or URL",
-      "claim": "What this evidence proves"
-    }
-  ],
-  "confidence": "confirmed|likely|hypothesis",
-  "severity": "critical|high|medium|low|info"
-}
-```
-
-## Action Candidate
-
-```json
-{
-  "sourceSkill": "seo-serp-opportunity-research",
-  "opportunityType": "missing_page|page_type_mismatch|content_gap|technical_fix|internal_link|ecommerce_enrichment",
-  "target": {},
-  "recommendedAction": "Specific action",
-  "expectedOutcome": "Observable improvement",
-  "impact": "high|medium|low",
-  "effort": "high|medium|low",
-  "priority": "critical|high|medium|low",
-  "confidence": "confirmed|likely|hypothesis",
-  "canAutoApply": false,
-  "requiresThirdPartyExecution": false,
-  "verificationMethod": "How to verify after apply",
-  "evidence": [],
-  "actionFingerprint": "sha256..."
-}
-```
-
-## Rules
-
-- `confirmed`: direct evidence proves the condition.
-- `likely`: multiple indirect signals support it.
-- `hypothesis`: plausible but needs more evidence before execution.
-- `canAutoApply=true` only when Anymorph can perform the mutation through CMS,
-  page generation, or managed action tooling.
-- Third-party tasks must route through third-party execution planning.
-- Duplicate actions share the same normalized target, operation, and payload
-  fingerprint and should be merged.
-

package/dist/skillpacks/geo/skills/seo-opportunity-audit/scripts/seo-toolkit.mjs

@@ -1,248 +0,0 @@
-#!/usr/bin/env node
-import { createHash } from "node:crypto";
-import { mkdir, readFile, writeFile } from "node:fs/promises";
-import { dirname, join, resolve } from "node:path";
-
-const command = process.argv[2];
-const args = parseArgs(process.argv.slice(3));
-
-try {
-	const result = await run(command, args);
-	if (result !== undefined) console.log(JSON.stringify(result, null, 2));
-} catch (err) {
-	const message = err instanceof Error ? err.message : String(err);
-	console.error(message);
-	process.exit(1);
-}
-
-async function run(mode, input) {
-	if (mode === "init-artifact") return initArtifact(input);
-	if (mode === "write-json") return writeJson(input);
-	if (mode === "fingerprint") return fingerprint(input);
-	if (mode === "dfs-request") return dfsRequest(input);
-	if (mode === "dfs-call") return dfsCall(input);
-	if (mode === "dfs-normalize") return dfsNormalize(input);
-	if (mode === "validate-artifact") return validateArtifact(input);
-	throw new Error("Usage: seo-toolkit.mjs <init-artifact|write-json|fingerprint|dfs-request|dfs-normalize|validate-artifact>");
-}
-
-function parseArgs(argv) {
-	const out = { repo: "." };
-	for (let i = 0; i < argv.length; i += 1) {
-		const arg = argv[i];
-		if (arg === "--skill") out.skill = requireValue(argv, ++i, arg);
-		else if (arg === "--run-id") out.runId = requireValue(argv, ++i, arg);
-		else if (arg === "--repo") out.repo = requireValue(argv, ++i, arg);
-		else if (arg === "--name") out.name = requireValue(argv, ++i, arg);
-		else if (arg === "--file") out.file = requireValue(argv, ++i, arg);
-		else if (arg === "--json") out.json = requireValue(argv, ++i, arg);
-		else if (arg === "--endpoint") out.endpoint = requireValue(argv, ++i, arg);
-		else if (arg === "--params") out.params = requireValue(argv, ++i, arg);
-		else if (arg === "--path") out.path = requireValue(argv, ++i, arg);
-		else if (arg === "--payload") out.payload = requireValue(argv, ++i, arg);
-		else throw new Error(`Unknown argument: ${arg}`);
-	}
-	return out;
-}
-
-function requireValue(argv, index, flag) {
-	const value = argv[index];
-	if (!value || value.startsWith("--")) throw new Error(`${flag} requires a value`);
-	return value;
-}
-
-async function initArtifact(input) {
-	const skill = required(input.skill, "--skill");
-	const runId = required(input.runId, "--run-id");
-	const repo = resolve(input.repo ?? ".");
-	const dir = join(repo, "agent", "seo", skill, runId);
-	await mkdir(dir, { recursive: true });
-	const manifest = {
-		skill,
-		runId,
-		createdAt: new Date().toISOString(),
-		files: [],
-	};
-	await writeFile(join(dir, "manifest.json"), `${JSON.stringify(manifest, null, 2)}\n`);
-	return { ok: true, dir };
-}
-
-async function writeJson(input) {
-	const skill = required(input.skill, "--skill");
-	const runId = required(input.runId, "--run-id");
-	const name = required(input.name, "--name");
-	const repo = resolve(input.repo ?? ".");
-	const dir = join(repo, "agent", "seo", skill, runId);
-	await mkdir(dir, { recursive: true });
-	const payload = input.json ? JSON.parse(input.json) : JSON.parse(await readStdin());
-	const path = join(dir, name.endsWith(".json") ? name : `${name}.json`);
-	await writeFile(path, `${JSON.stringify(payload, null, 2)}\n`);
-	return { ok: true, path };
-}
-
-function fingerprint(input) {
-	const payload = input.json ? JSON.parse(input.json) : input;
-	const canonical = stableStringify(payload);
-	return {
-		fingerprint: createHash("sha256").update(canonical).digest("hex"),
-		canonical,
-	};
-}
-
-function dfsRequest(input) {
-	const endpoint = required(input.endpoint, "--endpoint");
-	const params = input.params ? JSON.parse(input.params) : {};
-	const cost = estimateDfsCost(endpoint, params);
-	return {
-		endpoint,
-		mcpTool: endpointToMcpTool(endpoint),
-		params,
-		estimatedCostUsd: cost,
-		approvalStatus: cost >= 0.25 ? "needs_approval" : "approved",
-		notes: [
-			"Save this request next to the skill artifact before calling DFS.",
-			"After the MCP/API call, pipe the raw response through dfs-normalize.",
-		],
-	};
-}
-
-async function dfsCall(input) {
-	const apiPath = required(input.path, "--path");
-	const payload = input.payload ? JSON.parse(input.payload) : JSON.parse(await readStdin());
-	const login = process.env.DATAFORSEO_LOGIN || process.env.DATAFORSEO_USERNAME;
-	const password = process.env.DATAFORSEO_PASSWORD;
-	if (!login || !password) {
-		throw new Error("DATAFORSEO_LOGIN/DATAFORSEO_USERNAME and DATAFORSEO_PASSWORD are required for dfs-call");
-	}
-	if (!apiPath.startsWith("/v3/")) throw new Error("--path must start with /v3/");
-	const auth = Buffer.from(`${login}:${password}`).toString("base64");
-	const response = await fetch(`https://api.dataforseo.com${apiPath}`, {
-		method: "POST",
-		headers: {
-			Authorization: `Basic ${auth}`,
-			"Content-Type": "application/json",
-		},
-		body: JSON.stringify(payload),
-	});
-	const body = await response.text();
-	if (!response.ok) {
-		throw new Error(`DataForSEO ${response.status}: ${body.slice(0, 1000)}`);
-	}
-	return JSON.parse(body);
-}
-
-async function dfsNormalize(input) {
-	const endpoint = required(input.endpoint, "--endpoint");
-	const raw = JSON.parse(await readStdin());
-	return {
-		endpoint,
-		normalizedAt: new Date().toISOString(),
-		items: extractItems(raw).map((item) => normalizeDfsItem(endpoint, item)),
-		rawTaskStatus: extractStatus(raw),
-	};
-}
-
-async function validateArtifact(input) {
-	const file = required(input.file, "--file");
-	const parsed = JSON.parse(await readFile(file, "utf8"));
-	const problems = [];
-	if (!parsed || typeof parsed !== "object" || Array.isArray(parsed)) problems.push("artifact must be an object");
-	if (!parsed.runId && !parsed.skill && !parsed.actions && !parsed.findings && !parsed.opportunities) {
-		problems.push("artifact should include runId/skill or findings/opportunities/actions");
-	}
-	return { ok: problems.length === 0, file, problems };
-}
-
-function required(value, name) {
-	if (!value) throw new Error(`${name} is required`);
-	return value;
-}
-
-async function readStdin() {
-	const chunks = [];
-	for await (const chunk of process.stdin) chunks.push(chunk);
-	return Buffer.concat(chunks).toString("utf8");
-}
-
-function stableStringify(value) {
-	if (Array.isArray(value)) return `[${value.map(stableStringify).join(",")}]`;
-	if (value && typeof value === "object") {
-		return `{${Object.keys(value)
-			.sort()
-			.map((key) => `${JSON.stringify(key)}:${stableStringify(value[key])}`)
-			.join(",")}}`;
-	}
-	return JSON.stringify(value);
-}
-
-function estimateDfsCost(endpoint, params) {
-	const count =
-		Number(params.count) ||
-		(params.keywords && Array.isArray(params.keywords) ? params.keywords.length : 1) ||
-		1;
-	if (endpoint.includes("serp")) return Math.ceil(count / 100) * 0.002;
-	if (endpoint.includes("keyword") || endpoint.includes("related")) return Math.max(0.005, count * 0.0001);
-	if (endpoint.includes("competitor") || endpoint.includes("intersection")) return 0.01;
-	if (endpoint.includes("on_page") || endpoint.includes("lighthouse")) return 0.01;
-	return 0.005;
-}
-
-function endpointToMcpTool(endpoint) {
-	const aliases = {
-		serp_organic_live_advanced: "mcp__dataforseo__serp_organic_live_advanced",
-		dataforseo_labs_google_related_keywords: "mcp__dataforseo__dataforseo_labs_google_related_keywords",
-		dataforseo_labs_google_keyword_overview: "mcp__dataforseo__dataforseo_labs_google_keyword_overview",
-		dataforseo_labs_google_competitors_domain: "mcp__dataforseo__dataforseo_labs_google_competitors_domain",
-		dataforseo_labs_google_domain_intersection: "mcp__dataforseo__dataforseo_labs_google_domain_intersection",
-		dataforseo_labs_google_page_intersection: "mcp__dataforseo__dataforseo_labs_google_page_intersection",
-		dataforseo_labs_google_serp_competitors: "mcp__dataforseo__dataforseo_labs_google_serp_competitors",
-		anymorph_dataforseo_research: "mcp__anymorph__dataforseo_research",
-		anymorph_search_keywords: "mcp__anymorph__search_keywords",
-		anymorph_serp_snapshot: "mcp__anymorph__serp_snapshot",
-		anymorph_list_seo_opportunities: "mcp__anymorph__list_seo_opportunities",
-	};
-	return aliases[endpoint] ?? `mcp__dataforseo__${endpoint}`;
-}
-
-function extractItems(raw) {
-	if (Array.isArray(raw)) return raw;
-	if (Array.isArray(raw.items)) return raw.items;
-	if (Array.isArray(raw.result)) return raw.result.flatMap((r) => r.items ?? r.result ?? r);
-	if (Array.isArray(raw.tasks)) return raw.tasks.flatMap((task) => task.result ?? task.items ?? []);
-	if (raw.data && Array.isArray(raw.data.items)) return raw.data.items;
-	return [raw];
-}
-
-function extractStatus(raw) {
-	if (raw.status_message || raw.status_code) {
-		return { statusCode: raw.status_code, statusMessage: raw.status_message };
-	}
-	if (Array.isArray(raw.tasks)) {
-		return raw.tasks.map((task) => ({
-			statusCode: task.status_code,
-			statusMessage: task.status_message,
-		}));
-	}
-	return null;
-}
-
-function normalizeDfsItem(endpoint, item) {
-	const keyword = item.keyword ?? item.keyword_data?.keyword ?? item.keyword_info?.keyword ?? item.query;
-	const keywordInfo = item.keyword_info ?? item.keyword_data?.keyword_info ?? {};
-	const serp = item.serp_item ?? item.first_domain_serp_element ?? item.second_domain_serp_element ?? item;
-	return {
-		endpoint,
-		keyword,
-		searchVolume: keywordInfo.search_volume ?? item.search_volume ?? null,
-		cpc: keywordInfo.cpc ?? item.cpc ?? null,
-		competition: keywordInfo.competition ?? keywordInfo.competition_level ?? item.competition ?? null,
-		intent: item.search_intent_info?.main_intent ?? item.intent ?? null,
-		position: serp.rank_absolute ?? serp.position ?? serp.avg_position ?? item.avg_position ?? null,
-		url: serp.url ?? item.url ?? null,
-		domain: serp.domain ?? item.domain ?? null,
-		title: serp.title ?? item.title ?? null,
-		description: serp.description ?? item.description ?? null,
-		serpFeatures: item.serp_features ?? serp.serp_features ?? [],
-		raw: item,
-	};
-}

package/dist/skillpacks/geo/skills/seo-page-diagnosis/SKILL.md

@@ -1,56 +0,0 @@
----
-name: seo-page-diagnosis
-description: Use when diagnosing one or more pages for title, meta description, headings, content depth, page intent fit, schema, images, internal links, CTAs, or page-level SEO action candidates.
----
-
-# SEO Page Diagnosis
-
-Use this skill for page-level SEO diagnosis. It is not a site infrastructure
-audit. It inspects the page as a search result candidate and conversion surface.
-
-Follow the `claude-seo` single-page pattern: on-page SEO, content quality,
-technical page elements, schema, images, and practical recommendations.
-
-## Artifact
-
-Write outputs under:
-
-```text
-agent/seo/seo-page-diagnosis/{runId}/
-```
-
-Initialize with the shared toolkit from `$seo-opportunity-audit`.
-
-## References
-
-- `references/page-checks.md` for page scoring dimensions.
-- `../seo-opportunity-audit/references/action-contract.md` for output shape.
-
-## Workflow
-
-1. Select target pages from run context, candidate URLs, weak intents, or SERP
-   opportunity outputs.
-2. Extract title, meta description, H1, heading hierarchy, canonical, robots,
-   schema, image alt/dimensions, internal links, external links, CTA structure,
-   word count, and freshness signals.
-3. Compare page format with the intended query or intent if provided.
-4. Produce page-specific actions such as title rewrite, meta rewrite, section
-   addition, FAQ-like answer block, proof block, image alt update, schema
-   completion, or internal link placement.
-5. Separate auto-applicable CMS edits from third-party or source-code-only fixes.
-
-## Output Files
-
-- `pages.json`: page evidence and scorecards.
-- `findings.json`: findings by page.
-- `actions.json`: page-level action candidates.
-- `rationale.md`: concise explanation.
-
-## Guardrails
-
-- Do not claim ranking opportunity without SERP/DFS evidence.
-- Do not recommend keyword stuffing.
-- Do not recommend FAQ schema for Google rich result value unless the site type
-  qualifies; FAQ-like content can still help AI answerability.
-- If rendered content is unavailable, label findings as limited.
-

package/dist/skillpacks/geo/skills/seo-page-diagnosis/agents/openai.yaml

@@ -1,5 +0,0 @@
-interface:
-  display_name: "SEO Page Diagnosis"
-  short_description: "Diagnose page-level SEO and content action candidates."
-  default_prompt: "Use $seo-page-diagnosis to diagnose title, meta, headings, content depth, schema, images, and page-level SEO opportunities."
-

package/dist/skillpacks/geo/skills/seo-page-diagnosis/references/page-checks.md

@@ -1,38 +0,0 @@
-# Page Checks
-
-## On-Page SEO
-
-- title is unique, specific, and aligned with primary intent.
-- meta description summarizes value and supports click intent.
-- exactly one clear H1 in most cases.
-- headings form a logical outline without skipped meaning.
-- URL is short, stable, and descriptive.
-- canonical and robots directives match the desired index state.
-
-## Content Quality
-
-- page answers the primary question directly.
-- content depth matches page type.
-- claims include proof, examples, data, screenshots, customers, or source links.
-- author, brand, or product expertise is visible where relevant.
-- stale content has last-updated or freshness risk.
-
-## AI Answerability
-
-- include self-contained answer blocks.
-- define entities clearly.
-- state comparisons, criteria, pros/cons, and facts in extractable language.
-- avoid vague marketing-only claims.
-
-## Images and Media
-
-- important images have descriptive alt text.
-- large hero/product images have dimensions and modern formats where possible.
-- visual evidence supports the page purpose.
-
-## Links
-
-- internal links point to next useful pages.
-- anchors are descriptive and natural.
-- external links support factual claims when needed.
-

package/dist/skillpacks/geo/skills/seo-serp-opportunity-research/SKILL.md

@@ -1,66 +0,0 @@
----
-name: seo-serp-opportunity-research
-description: Use when researching SEO opportunities from DataForSEO SERPs, keyword gaps, SERP overlap, page-type mismatch, competitor pages, query clusters, or new page opportunities.
----
-
-# SEO SERP Opportunity Research
-
-Use this skill to find search opportunities that require live SERP or keyword
-evidence. It combines `claude-seo` patterns from `seo-cluster`, `seo-sxo`,
-`seo-dataforseo`, and competitor-page guidance.
-
-DFS is the primary data source. Firecrawl is only for reading specific pages
-after DFS identifies competitors or candidate URLs.
-
-## Artifact
-
-Write outputs under:
-
-```text
-agent/seo/seo-serp-opportunity-research/{runId}/
-```
-
-Use the shared toolkit for DFS request artifacts and normalization.
-
-## References
-
-- `references/serp-methodology.md` for DFS/SERP workflow.
-- `references/page-type-taxonomy.md` for SXO classification.
-- `../seo-opportunity-audit/references/action-contract.md` for output shape.
-
-## Workflow
-
-1. Gather seed terms from intents, workspace category, product catalog, existing
-   pages, GSC/SEO opportunity MCP results, and user input.
-2. Use the SEO toolkit to create DFS request artifacts for related keywords,
-   keyword overview, SERP competitors, ranked/intersection/page intersection, or
-   organic SERP snapshots.
-3. Normalize DFS responses before analysis.
-4. Cluster queries by SERP overlap and intent.
-5. Classify the dominant page type for each cluster.
-6. Detect opportunities:
-   - missing page.
-   - page-type mismatch.
-   - content gap.
-   - comparison/alternatives/best-of opportunity.
-   - weak existing page needing refresh.
-   - internal-link support opportunity.
-7. Use Firecrawl to inspect only the top competitor pages needed for evidence.
-8. Emit opportunities and action candidates.
-
-## Output Files
-
-- `dfs-requests.json`: planned DFS calls and cost tiers.
-- `dfs-normalized.json`: normalized DFS evidence.
-- `clusters.json`: query clusters and page-type consensus.
-- `opportunities.json`: ranked opportunities.
-- `actions.json`: action candidates.
-- `rationale.md`: concise decision record.
-
-## Confidence Rules
-
-- `confirmed`: DFS SERP/keyword evidence and page inventory agree.
-- `likely`: DFS evidence is strong but page inventory or competitor scrape is
-  incomplete.
-- `hypothesis`: keyword or SERP signal exists but needs validation before action.
-

package/dist/skillpacks/geo/skills/seo-serp-opportunity-research/agents/openai.yaml

@@ -1,5 +0,0 @@
-interface:
-  display_name: "SEO SERP Opportunity Research"
-  short_description: "Find DFS-backed keyword, SERP, and page-type opportunities."
-  default_prompt: "Use $seo-serp-opportunity-research to find DFS-backed SERP, keyword, cluster, and competitor page opportunities."
-

package/dist/skillpacks/geo/skills/seo-serp-opportunity-research/references/page-type-taxonomy.md

@@ -1,40 +0,0 @@
-# Page-Type Taxonomy
-
-Use this for SXO-style page-type mismatch detection.
-
-## Common Types
-
-- `homepage`
-- `product`
-- `pricing`
-- `feature`
-- `integration`
-- `comparison`
-- `alternatives`
-- `best_of`
-- `category`
-- `collection`
-- `product_detail`
-- `guide`
-- `how_to`
-- `glossary`
-- `case_study`
-- `review`
-- `tool`
-- `calculator`
-- `directory`
-- `local_service`
-
-## Mismatch Examples
-
-- SERP shows product/category pages, target is a blog post: high mismatch.
-- SERP shows comparison/listicle pages, target is a product page: high mismatch.
-- SERP shows local pack/service pages, target lacks local signals: medium/high.
-- SERP is fragmented: opportunity for differentiation, not a hard failure.
-
-## Consensus
-
-- Strong consensus: one type is above 60% of top organic results.
-- Mixed consensus: top type is 40-60%.
-- Fragmented: no type reaches 40%.
-

package/dist/skillpacks/geo/skills/seo-serp-opportunity-research/references/serp-methodology.md

@@ -1,38 +0,0 @@
-# SERP Opportunity Methodology
-
-## DFS-First Data Flow
-
-1. Expand seeds with related keywords and keyword overview.
-2. Fetch SERP competitors or organic SERP snapshots for candidate clusters.
-3. Use domain/page intersection to find competitor keywords the workspace lacks.
-4. Normalize every DFS response before reasoning.
-5. Scrape competitor pages only after a cluster is worth inspecting.
-
-## SERP Overlap
-
-Group keywords by shared top results:
-
-- 7-10 shared URLs: same target page.
-- 4-6 shared URLs: same cluster.
-- 2-3 shared URLs: adjacent clusters with cross-links.
-- 0-1 shared URLs: separate opportunity.
-
-Use DFS SERP data when available. If DFS is unavailable, mark results as
-`hypothesis` and explain the limitation.
-
-## Opportunity Types
-
-- `missing_page`: no matching owned/GEO page exists for a valuable cluster.
-- `page_type_mismatch`: existing page type differs from SERP consensus.
-- `content_gap`: page exists but lacks expected sections/proof/schema/media.
-- `comparison_page_gap`: SERP expects vs/alternatives/best-of content.
-- `refresh_existing_page`: page exists but freshness/depth/intent is weak.
-- `internal_link_support`: page exists but needs stronger internal links.
-
-## Competitor Page Rules
-
-- Use public, verifiable competitor claims only.
-- Separate fact tables from interpretation.
-- Avoid unsupported negative competitor claims.
-- Include last-updated recommendation for comparison pages.
-

package/dist/skillpacks/geo/skills/seo-technical-diagnosis/SKILL.md

@@ -1,64 +0,0 @@
----
-name: seo-technical-diagnosis
-description: Use when diagnosing technical SEO blockers, crawlability, indexability, robots, sitemap, canonicals, rendering, Core Web Vitals, security headers, AI crawler access, or technical action candidates.
----
-
-# SEO Technical Diagnosis
-
-Use this skill for site and URL-level technical SEO evidence.
-
-This follows the `claude-seo` technical audit pattern: crawlability,
-indexability, security, URL structure, mobile, CWV, structured data, JavaScript
-rendering, IndexNow, and AI crawler management.
-
-## Artifact
-
-Write outputs under:
-
-```text
-agent/seo/seo-technical-diagnosis/{runId}/
-```
-
-Initialize with:
-
-```bash
-SEO_TOOLKIT=.agents/skills/seo-opportunity-audit/scripts/seo-toolkit.mjs
-test -f "$SEO_TOOLKIT" || SEO_TOOLKIT=.claude/skills/seo-opportunity-audit/scripts/seo-toolkit.mjs
-node "$SEO_TOOLKIT" init-artifact --skill seo-technical-diagnosis --run-id "$RUN_ID"
-```
-
-## References
-
-Read when needed:
-
-- `references/checks.md` for technical check categories and severity rules.
-- `../seo-opportunity-audit/references/action-contract.md` for output shape.
-
-## Workflow
-
-1. Start from known workspace URLs, sitemap URLs, crawl logs, and CMS inventory.
-2. Check robots, sitemap, status codes, redirects, canonical, robots meta,
-   hreflang, title/meta availability, schema presence, JS rendering risk, mobile,
-   CWV/PSI/DFS on-page data if available, and AI crawler access.
-3. Use DFS on-page/Lighthouse only through the SEO toolkit request/normalize flow.
-4. Classify each issue by indexing risk, ranking risk, AI visibility risk, and
-   ability to auto-apply.
-5. Produce `findings.json`, `actions.json`, `blocked.json`, and `rationale.md`.
-
-## Output Rules
-
-- `critical`: blocks indexing/crawling, severe canonical/noindex conflict, 5xx on
-  key pages, or AI/search crawler fully blocked when visibility is intended.
-- `high`: major templates missing canonical/schema, important pages JS-only, bad
-  redirect chains, poor CWV with field/lab evidence.
-- `medium`: optimization opportunity, partial crawler issue, incomplete sitemap,
-  missing schema on eligible pages.
-- `low`: polish or monitoring.
-
-Every action must include verification:
-
-- recrawl URL.
-- compare rendered and raw HTML.
-- validate schema.
-- rerun DFS/PSI/CWV check.
-- inspect CMS/source diff when auto-applied.

package/dist/skillpacks/geo/skills/seo-technical-diagnosis/agents/openai.yaml

@@ -1,5 +0,0 @@
-interface:
-  display_name: "SEO Technical Diagnosis"
-  short_description: "Diagnose technical SEO blockers and action candidates."
-  default_prompt: "Use $seo-technical-diagnosis to diagnose crawlability, indexability, rendering, schema, and technical SEO issues."
-